NER_DICTIONARY_DIMENSION_PROB=0.65
NER_DICTIONARY_MEASURE_PROB=0.65
NER_DICTIONARY_CONSTANTENTITIES_PROB=0.65


reports.nlp.can.use.data=true
reports.nlp.can.use.data.autosuggestvalidation=true


RAZ_VISUALIZATION_SUGGESTIONS_FILE_PATH=./conf/nlp/dictionary/RAZ/ENGLISH/autosuggest/en-raz-visualizationPriority.xml
RAZ_CONCEPTS_ALLOWED_ENTITIES_FILE_PATH=./conf/nlp/dictionary/RAZ/conceptEntities.xml
RAZ_OPENNLP_SENTENCE_PATTERN_FILE_PATH = ./conf/nlp/dictionary/RAZ/sentence_patterns.txt


RANKING_ENGINE_MULTITHREADFLOW_ENABLED=true

MAPPERENGINE_CLASS=com.zoho.nlp.raz.MapperEngineRelImpl
COMBINATION_VALIDATOR_IMPL=com.zoho.nlp.raz.CombinationValidatorDefaultImpl
    
USE_DEFAULT_VALUES_FOR_BOOLEAN=true
DEFAULT_BOOLEAN_COLUMN_TRUE_VALUE=Yes
DEFAULT_BOOLEAN_COLUMN_FALSE_VALUE=No

RAZ_ENTITIES_REORDERING_ALLOWED=true
RAZ_ALLOW_PARTIAL_MATCHES=true

MAX_COLUMNS_START_END_TIME=3

NLP_COLUMN_DICTIONARY_APPROACH=trie
SPELLING_CORRECTION_MINIMUM_INITIAL_CHARACTERS=2
    
CORPORA_CONNECTOR_WORDS_FILE=./conf/nlp/dictionary/misc/corpora_connector_words.txt

IS_WORD_SPLITTER_ENABLED=true

DICTIONARY_THRESHOLD_PROB=0.3
RAZ_ENTITIES_PROCESSING_ITERATIONS_COUNT=1

SUPPORTED_LANGUAGES=ENGLISH,SPANISH,FRENCH,PORTUGUESE

NLPCORE_CLASS=com.zoho.nlp.raz.RAZConversationCoreImpl

CONVERSATION_NLPCORE_CLASS=com.zoho.nlp.raz.RAZConversationCoreImpl

EXTRACTION_ENGINE_CLASS={"DEPENDENCY_PARSER_IMPL":"com.zoho.nlp.raz.SDPCoreImpl", "SAZ_EXTRACTION_IMPL":"com.zoho.nlp.raz.SAZCoreImpl", "OPENAI_EXTRACTION_IMPL":"com.zoho.nlp.raz.OpenAICoreImpl"}

NLP_INTERPRETER_CLASS=com.zoho.nlp.raz.interpreter.RAZInterpreter
NLP_RANKER_CLASS=com.zoho.nlp.raz.ranking.RAZRanker
NLP_FUNCTION_CLASS=com.zoho.nlp.raz.function.RAZFunction
NLP_CONVERSATION_CLASS=com.zoho.nlp.raz.conversation.RAZConversation
NLP_OPENAI_CONVERSATION_CLASS=com.zoho.nlp.raz.conversation.OpenAIConversation
NLPINTERPRETATION_CLASS=com.zoho.nlp.raz.interpretation.RAZInterpreterAPI
ASENGINE_API_CLASS=com.zoho.nlp.raz.autosuggestion.RAZASEngineAPI
PHRASEPREDICTOR_CONFIGURATIONS_CLASS=com.zoho.nlp.raz.autosuggestion.phrasepredictor.RAZPhrasePredictorConfigurations
NLP_SERVICE_PROPERTY_FETCHER_CLASS=com.zoho.nlp.raz.util.RAZPropertyFetcher
 
UNSTRUCTURED_QUERY_SUPPORT=com.zoho.nlp.raz.unstructured.RAZUnStrQrySup
QUERY_DETAILS_EXTRACTOR_CLASS=com.zoho.nlp.raz.querydetails.RAZQueryDetailsExtractor
MAPPER_CLASS=com.zoho.nlp.raz.mapper.RAZMapper
NLP_MAPPER_CLASS=com.zoho.nlp.raz.demo.xml.XMLRAZMapper
NLP_PERMISSION_CLASS=com.zoho.nlp.raz.demo.xml.XMLRAZDMPermission
ENTITYINFO_CLASS=com.zoho.nlp.raz.mapper.RAZEntityInfo
COMBINATION_VALIDATOR_IMPL=com.adventnet.zoho.db.nlp.ZRNLPCombinationValidator
RANKINGENGINE_CLASS=com.zoho.nlp.raz.ranking.RAZRankingEngine
INTERPRETER_CLASS=com.zoho.nlp.raz.interpreter.RAZInterpreter
ENTITYMATCHPOPULATOR_CLASS=com.zoho.nlp.raz.matchfinder.RAZEntityMatchPopulator
NER_DICT_MATCHER_CLASS=com.zoho.nlp.raz.datamodel.dict.RAZNERDictMatcher
NER_INTERPRETER_CLASS=com.zoho.nlp.raz.ner.RAZNERInterpreter    
RESOLVECONFUSION_CLASS=com.zoho.nlp.raz.confusion.RAZResolveConfusionsImpl
DATAMODELAPI_CLASS=com.zoho.nlp.raz.trie.dictionary.dbmeta.RAZDBMetaImpl
GEODATA_API_CLASS=com.zoho.nlp.datamodel.dict.geodict.NLPGeoDataImpl
NLP_SUGGESTED_SYNONYMS_CLASS=com.zoho.nlp.raz.gptintegration.synonymsgenerator.SuggestSynonymsWithGPT

ASKZIA_ZLABS_AGENT_CLASS=com.zoho.nlp.raz.ZLabsAgentEngine
ASKZIA_ZLABS_AGENT_OUTPUT_BUILDER_CLASS=com.zoho.nlp.raz.agent.ReportCreationAgentEngine
 
NLPDATAMODEL_CLASS=com.adventnet.zoho.db.nlp.ZRNLPDataModel
RANKCOLLECTOR_CLASS=com.zoho.nlp.ranking.raz.RAZRankCollector
SERVICEMODELLOADER_CLASS=com.zoho.nlp.raz.resource.models.RAZServiceModelLoader
RAZ_ALLOW_ADVANCED_UNSTRUCTURED_TO_STRUCTURED_IMPL=2

NLP_DM_META_CLASS=com.zoho.nlp.raz.demo.xml.XMLRAZDMMeta
NLP_DM_DATA_CLASS=com.zoho.nlp.raz.demo.xml.XMLRAZDMData
NLP_DM_DATA_MATCHFINDER_WRAPPER_CLASS=com.zoho.nlp.analytics.datamatchfinder.AnalyticsColumnDataAPI

NLP_TRASH_HANDLER=com.zoho.nlp.raz.thrashhandler.RAZTrashedEntitiesHandler

DEMO_DM_XML_PATH=datamodel.xml
SPANISH_DEMO_DM_XML_PATH=datamodel.xml
NLP_IS_CONVERSATION_ENABLED=true

LANGUAGE_DETECTION_CLASS=com.zoho.nlp.raz.machinetranslation.langdetect.JSONLanguageDetection
MACHINE_TRANSLATION_CLASS=com.zoho.nlp.raz.machinetranslation.RAZMachineTranslationEngine
NLP_DEFAULT_MERGE_SPAN_IMPLEMENTATION_FOR_SERVICE=com.zoho.nlp.ner.merge.DefaultMergeSpansImpl
NLP_NER_MERGE_SPAN_IMPLEMENTATION_FOR_SERVICE=com.zoho.nlp.ner.merge.NERMergeSpanImplBasedOnMatchCategory

RAZ_OPENAI_QD_EXTRACTOR_CLASS=com.adventnet.zoho.db.nlp.OpenAIQueryDetailsExtractorAskZia
RAZ_OPENAI_COMPLEX_QUERY_EXTRACTOR_CLASS=com.adventnet.zoho.db.nlp.OpenAIComplexQueryDetailsExtractorAskZia
RAZ_OPENAI_SKILL_IDENTIFIER_CLASS=com.adventnet.zoho.db.nlp.OpenAISkillClassifierAskZia
RAZ_OPENAI_ACTION_SKILL_IDENTIFIER_CLASS=com.adventnet.zoho.db.nlp.OpenAIActionSkillClassifierAskZia
RAZ_OPENAI_SQL_AND_FORMULA_SKILL_IDENTIFIER_CLASS=com.adventnet.zoho.db.nlp.OpenAISQLFormulaSkillClassifierAskZia
RAZ_OPENAI_ZIA_ACTION_SKILL_IDENTIFIER_CLASS=com.adventnet.zoho.db.nlp.OpenAIZIAActionSkillClassifierAskZia
RAZ_OPENAI_SLIDESHOW_ACTION_SKILL_IDENTIFIER_CLASS=com.adventnet.zoho.db.nlp.OpenAISlideshowSkillClassifierAskZia
RAZ_OPENAI_DASHBOARD_ACTION_SKILL_IDENTIFIER_CLASS=com.adventnet.zoho.db.nlp.OpenAIDashBoardSkillClassifierAskZia

GRAPH_PREDICTOR_MODEL=[{"MODEL_INDEX":GRAPH_PREDICTOR_MODEL,"MODEL_FILEPATH":"./conf/nlp/models/RAZ/zreports/dl-bestgraph-model"}]
GRAPHPREDICTOR_CLASS=com.zoho.nlp.raz.graphtypeprediction.dlmodel.MLBasedGraphTypePredictor
NLP_MAPPER_COMBINATION_THRESHOLD=1000
MAPPER_CLUSTER_COMBINATION_LIMIT=100
NLPTRACER_ACCESS_USERS=sridharks@zohocorp.com,search-team@zohocorp.com,rakesh.subrahmanyam@zohocorp.com,vinayakaram.n@zohocorp.com,sandeep.kotra@zohocorp.com,syed.kamarudheen@zohocorp.com,manimala@zohocorp.com,vijay.selvaraj@zohocorp.com,dineshkumar.g@zohocorp.com,nlp-team+1@zohocorp.com,jagan.e@zohocorp.com,abinesh.b@zohocorp.com,rohit.mv@zohocorp.com,dinesh.ravindran@zohotest.com,vijayadhithyan.mn@zohotest.com,lakshmi.prabha@zohotest.com,kasyap.kv@zohotest.com,manoj.gattam@zohotest.com,abuthahir.anwar@zohotest.com,vishal.ram@zohotest.com,harihara.p@zohotest.com,gopika.bm@zohotest.com

RAZ_INSIGHTS_TEMPLATES_FILE_PATH=./conf/nlp/dictionary/raz/Insights/InsightsNarrationTemplates.xml
DIAGNOSTIC_INSIGHTS_TEMPLATES_FILE_PATH=./conf/nlp/dictionary/raz/Insights/DiagnosticInsightsNarrationTemplates.xml
RAZ_INSIGHTS_PICTORIAL_REPRESENTATION_FILE_PATH=./conf/nlp/dictionary/RAZ/ENGLISH/Insights/InsightsPictorialRepresentation.xml

RAZ_QUERY_NEXT_TEMPLATE_CATEGORY_FILE_PATH=./conf/nlp/dictionary/RAZ/QuerySuggestion/next-template-category.xml


RAZ_CHART_NARRATION_TEMPLATES_XML=./conf/nlp/dictionary/RAZ/Narration/en-raz-chart_narration_templates.xml

RAZ_APPPROPERTY_ASK_ZIA_NARRATION_IMPL=com.zoho.nlp.raz.narration.RAZTemplatesNarrationEngine

RAZ_EXPRESSION_MODEL=./conf/nlp/models/RAZ/ENGLISH/zreports/en-expression-detector.bin
RAZ_EXPRESSION_LOG_FILE=./conf/nlp/expression_sentences_log.txt
RAZ_SPACY_SERVER_URL=http://spacynlp.localzoho.com

APPPROPERTY_RAZ_INSIGHTS_FORECAST_API=com.adventnet.zoho.db.nlp.insights.ZAInsightsForecastAPI
APPPROPERTY_RAZ_INSIGHTS_OUTLIER_API=com.adventnet.zoho.db.nlp.insights.ZAInsightsOutlierAPI

APPPROPERTY_INSIGHTS_SHOW_GROUP_INFO=true

RAZ_ALLOW_BLACKLISTED_TABLES_FOR_RELATIONSHIP=true

RAZ_PROPERTY_OPENAI_SYNONYMS_TEXT_API_PARAM_MODEL=text-davinci-003
RAZ_PROPERTY_OPENAI_SYNONYMS_TEXT_API_PARAM_TEMPERATURE=0.3
RAZ_PROPERTY_OPENAI_SYNONYMS_TEXT_API_PARAM_MAX_TOKENS=600

RAZ_PROPERTY_OPENAI_SYNONYMS_CHAT_API_PARAM_MODEL=gpt-3.5-turbo-1106
RAZ_PROPERTY_OPENAI_SYNONYMS_CHAT_API_PARAM_TEMPERATURE=0.3
RAZ_PROPERTY_OPENAI_SYNONYMS_CHAT_API_PARAM_MAX_TOKENS=3000

RAZ_PROPERTY_OPENAI_SYNONYMS_FEATURE_ID=4
RAZ_PROPERTY_OPENAI_SYNONYMS_COUNT=20
RAZ_PROPERTY_OPENAI_SYNONYMS_PROMPT=generate {0} contextual synonyms for the column "{1}" as comma-separated values
RAZ_PROPERTY_OPENAI_SYNONYMS_PROMPT_WITH_TABLE=generate {0} contextual synonyms for the column "{1}" in "{2}" table as comma-separated values
RAZ_PROPERTY_OPENAI_SYNONYMS_PROMPT_WITH_DOMAIN=generate {0} contextual synonyms for "{1}" in "{2}" domain as comma-separated values
RAZ_PROPERTY_OPENAI_SYNONYMS_PROMPT_WITH_LANG=generate {0} {2} language contextual synonyms for "{1}" as comma-separated values
RAZ_PROPERTY_OPENAI_SYNONYMS_PROMPT_WITH_LANG_DOMAIN=generate {0} {3} language contextual synonyms for "{1}" in "{2}" domain as comma-separated values
RAZ_PROPERTY_OPENAI_SYNONYMS_PROMPT_FOR_TABLE=generate {0} contextual synonyms for the table "{1}" as comma-separated values

STARTS_WITH_MATCHES_LIMIT=20
ENDS_WITH_MATCHES_LIMIT=20
CONTAINS_MATCHES_LIMIT=20
STARTS_WITH_CHARACTERS_LIMIT=3
ENDS_WITH_CHARACTERS_LIMIT=3
CONTAINS_CHARACTERS_LIMIT=3

IS_SECONDARY_IMPL_ENABLED_FOR_AUTOSUGGEST=false
USE_SECONDARY_IMPL_FOR_DICTIONARY_MATCHES=false
IS_SECONDARY_IMPL_ENABLED_FOR_QUERY_SUGGESTIONS=false

CONVERSATION_SKILL_CONFIG_FILE_PATH=./conf/nlp/dictionary/RAZ/ENGLISH/core/conversation_skill_configurations.xml

CONVERSATION_SENTENCE_LOG_FILE_PATH=./conf/nlp/conv_sentences_log.txt

OPENAI_FLOW_LOGS_FILE_PATH=./conf/nlp/openai_flow_log.txt

CQ_TESTCASE_LOGS_FILE_PATH=./conf/nlp/cq_testcase_log.txt

CONVERSATION_ACTION_CONFIG_FILE_PATH=./conf/nlp/dictionary/RAZ/ENGLISH/core/conversation_action_configurations.xml

CONVERSATION_ACTION_SKILL_MODULE_KEY_CONFIG_FILE_PATH=./conf/nlp/dictionary/RAZ/ENGLISH/core/conv_action_module_key_configurations.xml

CONVERSATION_ACTION_SKILL_EXTRACTION_TABLES_FILE_PATH=./conf/nlp/dictionary/RAZ/ENGLISH/core/tables.xml

CONVERSATION_ACTION_SKILL_EXTRACTION_OPENAI_IMPLEMENTATION=com.adventnet.zoho.db.nlp.ActionExtractionAskZia

APP_PROPERTY_DASH_ANALYST_CONTEXT_PROVIDER=com.adventnet.zoho.db.nlp.DashboardAnalystZiaContextProvider
ACTION_SKILL_MATCHES_HANDLER_CLASS=com.zoho.nlp.raz.conversation.skills.RAZActionSkillMatchesHandler

SERVICE_CONVERSATION_SKILL_CONFIGURATION_FILE=./conf/nlp/dictionary/RAZ/Skills/service_skill_configurations.xml
CONVERSATION_ACTION_SKILL_MODULES_XML=./conf/nlp/dictionary/RAZ/ENGLISH/core/action_skill_modules.xml

za.askzia.kda_column.prompt.for.zia.turbo=As a data analyst, first remove any columns containing Personally Identifiable Information (PII) such as names, email addresses, phone numbers, or personal identifiers from the schema to ensure privacy. Then, as a causal analyst, identify at least three categorical dimensions (e.g., product category, region) and up to three numerical measures (e.g., cost, order value) that can causally influence a user-defined target column, based on the provided data model. Exclude IDs, dates, emails, phone numbers, and logically invalid inputs (e.g., profit as cause of sales). Use your domain understanding to guide causal logic. Output strictly in JSON format only, using hyphens to denote causal direction. Format: {"query": "<user-defined sentence>","target_column": "<target column>","Suggested_columns": [{"table_name": "<table_name>", "column_name": "<column_name>"}],"info": "Causal graph: Column A - Target Column, Column B - Target Column. Domain: <Identified domain>"}

za.askzia.key_drivers.prompt.for.zia.turbo=You are an AI assistant specialized in causal analysis. Your task is to identify causal relationships between variables in datasets and explain the key drivers behind observed changes in a target metric. Use causal inference techniques such as causal graphs, treatment effect estimation, and counterfactual analysis to provide actionable insights.\n **Key Requirements:** \n 1. **Key Drivers Identification**: Only include columns and their specific values (categories) that have a measurable and statistically significant impact on the target column. Exclude columns or values with no significant impact.\n 2. **Strict JSON Output**: The output must strictly adhere to the following JSON structure without any additional text or explanation:\n{"diagnostic_insights": [{"KeyDrivers": [{"column": "<Column Name>","category": "<Key Driver Value of the Column>",\n"insights": [\n"<Quantitative explanation of why this value is a key driver, e.g., specific numbers, percentages, or metrics.>",\n"<Another detailed insight supported by quantitative evidence.>"]},{"column": "<Column Name>",\n"category": "<Key Driver Value of the Column>",\n"insights": [\n"<Insight with numerical or statistical validation>",\n"<Insight supported by metrics or evidence>"]}]}]}\n**Output Guidelines:** \n- **Column Name**: Represents the name of the key driver column. \n - **Category**: Refers to the specific value within the column that is a key driver. \n - **Insights**: A list of quantitative insights explaining why this column value (category) is a key driver. \n - Only include impactful columns and their values (categories) as key drivers. \n - Exclude any text, comments, or explanation outside of the JSON structure.

za.askzia.insight_tabular_summary.prompt.for.zia.turbo=You are analyzing datasets and summarizing insights. Follow these rules for output: 1. **Dimension-Based Insights**: If insights include dimensions (e.g., Region, Product Category), use a JSON table format with columns and rows. Example: {\"insight_summary\": {\"columns\": [\"Category\", \"Metric 1\", \"Metric 2\", \"Anomalies\"], \"rows\": [[\"Region A\", \"$1.2M\", \"25%\", \"Spike in Q4\"], [\"Region B\", \"$900K\", \"20%\", \"Stable\"]]}}. 2. **Non-Dimensional Insights**: For global or overall insights without categorical splits, summarize in a concise string. Example: {\"insight_summary\": \"Total Sales from 2020-2024 were $1.3M. Max Sales: $62K in Dec 2022. Min Sales: $6K in Oct 2022. Sales dropped 48% overall.\"}. 3. **How to Choose the Format**: If insights compare across dimensions, use the table format. If insights summarize overall trends, use a text summary. 4. **Examples**: Input with Dimensions: **Meta Columns**: Sales, Region **Insights**: - Region A: Sales increased 10% in Q4. - Region B: Sales declined 5%. Output: {\"insight_summary\": {\"columns\": [\"Region\", \"Sales Change\"], \"rows\": [[\"Region A\", \"10% increase in Q4\"], [\"Region B\", \"5% decline\"]]}}. Input without Dimensions: **Meta Columns**: Sales **Insights**: - Total Sales: $1.3M from 2020-2024. - Max Sales: $62K in Dec 2022. Output: {\"insight_summaries\": \"Total Sales from 2020-2024 were $1.3M. Max Sales: $62K in Dec 2022.\"}. 5. Always ensure outputs are structured and readable: Use meaningful column names for tables. Keep summaries concise and relevant.

za.askzia.insight_summary.prompt.for.zia.turbo=Summarize the provided insights into 2 to 3 concise, meaningful sentences, focusing on overall performance, key trends, and significant variations. Highlight major peaks, declines, and contributions from top categories while ensuring the summary is easy to read and impactful. The output should be a JSON structure as: {\"insight_summary\":\"<summary of insights>\"}.

za.askzia.follow_up_queries.prompt.for.zia.turbo=Your task is to generate structured follow-up queries based on the provided conversation history, datamodel, and generated insights. Each query must be relevant to the assigned category and organized in a specific JSON structure.\n\nCategories and Instructions:\n\n1. Dimension-Based Queries (RelatedQueries):\n   - Identify the best dimension (categorical column) from the datamodel or conversation history that explains the given metric of analysis.\n   - Generate three queries that break down the metric using these dimensions.\n   - Example format: \"Breakdown [metric] by [dimension].\"\n\n2. Insight-Based Short Queries (InsightQueries):\n   - Based on generated insights and the conversation history, create five short and direct questions.\n   - Focus on anomalies, changes, and projections derived from the insights.\n   - Ensure the questions are concise and designed for quick, data-driven answers.\n   - Example focus areas: unusual spikes, trends, drops, or future implications.\n\n3. Diagnostic Questions (DiagnosticQueries):\n   - Analyze the insights tagged as diagnosable and generate short diagnostic questions related to them.\n   - Example format:\n     - Diagnosable Insight: \"In 2023, total sales decreased by 85%.\"\n     - Diagnostic Questions: \"Why did sales decrease in 2023?\" or \"What factors contributed to the sales decline in 2023?\"\n\nOutput Format:\nReturn the generated queries in the following JSON structure:\n\n{\n  \"FollowUpQueries\": {\n    \"RelatedQueries\": [\n      {\"query\": \"Breakdown [metric] by [dimension].\"},\n      {\"query\": \"Another breakdown query.\"},\n      {\"query\": \"Third breakdown query.\"}\n    ],\n    \"InsightQueries\": [\n      {\"query\": \"Is there any outliers?\"},\n      {\"query\": \"Min and Max Sales\"},\n      {\"query\": \"Forecast metric for next three months\"},\n      {\"query\": \"Is there any patterns/trends observed\"}\n    ],\n    \"DiagnosticQueries\": [\n      {\"query\": \"Why did [metric] change in [year]?\"},\n      {\"query\": \"What factors caused [insight]?\"}\n    ]\n  }\n}\n\nContext Usage:\n- For RelatedQueries, use the datamodel or conversation context to identify the most relevant dimensions.\n- For InsightQueries, use the generated insights to craft direct and actionable questions.\n- For DiagnosticQueries, use diagnosable insights to create focused diagnostic follow-ups.\n\nEnsure all queries are contextually aligned and concise. Return only the JSON structure as output.

za.askzia.prescriptive_insights.prompt.for.zia.turbo=You are an advanced data analysis assistant specializing in prescriptive insights derived from numerical data. Your task is to generate precise, structured, and highly specific actionable recommendations based on numerical data while strictly ignoring numeric range columns. Each recommendation must specify: WHAT action to take WHERE to implement it HOW MUCH to adjust (in percentage values) Dynamically reference available numerical columns while ensuring that each JSON object contains at most one token key with its corresponding token-type value. General Guidelines for Recommendations 1. Data-Based Recommendations If a metric is stored as a range, do not attempt to infer an exact number. Instead of displaying raw ranges, convert them into qualitative insights; (Incorrect) \"Boutique Bonanza drove 20000 to 40000 conversions on Instagram Mobile.\" (Correct) \"Boutique Bonanza achieved a strong number of conversions on Instagram Mobile.\" (Correct)  \"Boutique Bonanza saw high engagement, making it a top-performing campaign.\" If necessary, mention that a metric was among the highest\/lowest without specifying the range. Use comparative terms such as \"outperformed other campaigns\" or \"saw significant engagement\" where applicable. Do not infer additional context or make assumptions beyond the given data. 2. Exclusion of Numeric Range Data Never display numeric range values in recommendations (e.g., \"20000 to 40000\"). Instead of showing exact ranges, describe the trend in qualitative terms. Use appropriate descriptive terms based on the range: If the range is very high (e.g., 80000 to 100000), use:\"exceptionally high conversions\" If the range is moderate (e.g., 20000 to 40000), use:\"a strong number of conversions\" If the range is low (e.g., 0 to 20000), use:\"relatively lower conversions\" Always ensure the recommendation is clear and action-oriented without numeric range values. 3. Data Relevance If the dataset does not contain specific metrics (e.g., engagement metrics, ROI), do not mention them. Focus only on available metrics and ensure that recommendations are supported by data. The rationale section should be fully backed by the dataset without hypothetical statements. 4. Recommendations Narration Examples: Reduce Ad Spend by 20% for Furniture as it incurred $20,000 in ad costs but generated no sales Reduce Shipping Costs by 19.67% for Furniture as high shipping fees account for a large portion of the cost structure. In the above examples, \"Ad Spend\", \"Shipping Fees\" are columns , \"Furniture\" is a data. So, these kind of recommendations should be represented in our json format which is mentioned below. 5. Always apply directional logic: if a negative metric (e.g., loss, cost, churn) increases, recommend actions to reduce it; if a positive metric (e.g., revenue, profit, conversions) decreases, recommend actions to increase it; never suggest actions that would worsen any metric—recommendations must logically improve the business outcome. Exclusion of Date: Do not include date values (e.g., \"June 2023\") in the recommendation section. Replace time-specific phrases with general trend-based language. If necessary, reference the trend instead of the specific time period. Example Transformations: (Incorrect) \"Boost ad spend for Amazon-India by 15% as it was a key influencer in Month-Over-Month Net Qty increase in June 2023.\" (Correct)  \"Boost ad spend for Amazon-India by 15% as it played a significant role in the recent Month-Over-Month Net Qty growth.\" Entity-Specific Recommendations 1. When to Increase or Reduce Effort\/Spend Reducing Spend or Effort: Identify the least efficient entity where a decrease is beneficial. If a metric has already decreased, the recommendation should encourage further optimization rather than reversal. Example: If shipping cost has decreased significantly in 2023, recommend: \"Continue reducing shipping costs to further optimize expenses.\" Increasing Spend or Effort: Identify the high-performing entity that will benefit from additional investment. If a metric (e.g., profit) has increased, suggest further investment to amplify growth. Example: If profit has increased significantly in 2024, recommend: \"Increase marketing investment for high-performing products to capitalize on growth and drive further profit.\" Narration Style and Clarity 1. Narrative Variation Maintain a clear and concise narration style while ensuring variation in phrasing. Avoid repetitive structures; recommendations should be engaging and easy to read. 2. Maintaining One Data Point per Token Each JSON object should contain only one data value per token (e.g., \"Instagram\" or \"Mobile\"). Do not combine multiple values in a single token (e.g., \"Instagram Mobile\" should be split into two objects). 3. Data-Backed Reasoning: Every recommendation must explicitly reference numerical trends from the dataset. Do not use vague phrases such as \"as it is a key influencer\" or \"is a significant contributor to\". Instead, quantify the influence using numerical changes, for example: (Correct) \"Increase Ad Spend on Tablet by 15% since it drove a 33.3% increase in Revenue.\" Here, Ad spend, Revenue are columns and Tablet is data. (Correct) \"Boost Ad Spend for Twitter by 12% as Hot Summer saw 22% higher conversions compared to other channels.\" Here, Ad Spend, channel, conversions are columns and Twitter, Hot Summer are data. Recommendation Guidelines For each recommendation: Specify the action (e.g., increase, reduce). State what to adjust (e.g., budget allocation, ad spend). Mention where to implement it (e.g., Product, Region). Quantify the change (e.g., percentage increase or decrease). Do not include date values (e.g., \"in December 2023\") in the recommendation section. Date references should only appear in the rationale section if necessary to support the insight. If a recommendation is influenced by a specific time period, describe the trend without explicitly stating the date. Example Transformations: (Incorrect) \"Boost ad spend for Twitter by 10% as it was a key influencer in increasing ad spend in December 2023.\" (Correct) \"Boost ad spend for Twitter by 10% as it contributed significantly to recent ad spend growth.\" Example of Improved Narration Style Instead of: \"Increase Ad spend for Boutique by 5% on Facebook.\" Use: \"Increase Ad spend for Boutique by 5% on Facebook since Boutique has 10% more conversions.\" Ensure that recommendations do not contain date references. If a time-based trend is important, mention it only in the rationale section. Example Correction: (Incorrect) \"Increase ad spend for Twitter by 10% as it drove higher engagement in December 2023.\" (Correct)  \"Increase ad spend for Twitter by 10% as it showed strong engagement growth.\"Avoid drastic column name changes (e.g., if the column name is \"Ad Spend\", do not change it to \"Investment\"). However, minor adjustments such as singular\/plural forms or partial names are acceptable (e.g., \"campaign name\" : \"campaigns\" or \"campaign\"). Ensure that recommendations do not contain date references. Instead of stating a specific time period, describe the trend using relative terms. Example Correction: (Incorrect) \"Increase ad spend for Amazon-India by 15% as it contributed to the highest Month-Over-Month Net Qty in June 2023.\" (Correct) \"Increase ad spend for Amazon-India by 15% as it played a significant role in the highest recent Month-Over-Month Net Qty growth.\" JSON Formatting Rules 1. Token-Type Values COLUMN_NAME : Represents column names from the dataset. KEY_TEXT_VALUE : Represents specific column values (e.g., region names, categories). (Do not use KEY_TEXT_VALUE for any other token except actual data.) NUMBER : Represents numerical values including percentages. POSITIVE_NUMBER : Represents currency values indicating an increase. NEGATIVE_NUMBER : Represents currency values indicating a decrease. DATE_HIGHLIGHT values (e.g., \"June 2023\") should never appear in the recommendation section. If a date reference is relevant, it should only appear in the rationale section. If a recommendation refers to a trend influenced by a specific date, rephrase it without mentioning the date explicitly. Example Transformations: (Incorrect) \"Increase ad spend for Amazon-India by 15% as it was a key influencer in driving the highest Month-Over-Month Net Qty in June 2023.\" (Correct)  \"Increase ad spend for Amazon-India by 15% as it was a key driver of recent Month-Over-Month Net Qty growth.\" 2. Maximum Tokens Per JSON Object Each JSON object must contain at most one column name or one data value. Example: \"Instagram Mobile\" contains two data values (\"Instagram\" and \"Mobile\") : Split into two separate objects. 3. Handling Partial Column Names or Data If a JSON object contains a partial column name (e.g., \"Spend\" instead of \"Ad Spend\"), it should still be treated as a separate JSON object. The same applies to partial data values. Example of JSON Object for Recommendations { \"prescriptive_insights\": [ { \"recommendation\": [ {\"token\": \"Reduce\"}, {\"token\": \"ad spend\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \"by\"}, {\"token\": \"20%\", \"token-type\": \"NEGATIVE_NUMBER\"}, {\"token\": \"for\"}, {\"token\": \"Furniture\", \"token-type\": \"KEY_TEXT_VALUE\"}, {\"token\": \"as it incurred\"}, {\"token\": \"$20,000\", \"token-type\": \"NEGATIVE_NUMBER\"}, {\"token\": \"in\"}, {\"token\": \"ad cost\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \"but generated no\"}, {\"token\": \"sales\", \"token-type\": \"COLUMN_NAME\"} ], \"rationale\": [ {\"token\": \"Furniture\", \"token-type\": \"KEY_TEXT_VALUE\"}, {\"token\": \"incurred\"}, {\"token\": \"$20,000\", \"token-type\": \"NEGATIVE_NUMBER\"}, {\"token\": \"in\"}, {\"token\": \"ad cost\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \"but did not generate any\"}, {\"token\": \"sales\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \", making it inefficient.\"} ], \"heading\": \"Cut Losses in Non-Performing Category\" }, { \"recommendation\": [ {\"token\": \"As high\"}, {\"token\": \"shipping cost\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \"account for a large portion of\"}, {\"token\": \"expenses\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \",\"}, {\"token\": \"reduce\"}, {\"token\": \"shipping cost\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \"by\"}, {\"token\": \"19.67%\", \"token-type\": \"NEGATIVE_NUMBER\"}, {\"token\": \"for\"}, {\"token\": \"Furniture\", \"token-type\": \"KEY_TEXT_VALUE\"} ], \"rationale\": [ {\"token\": \"Shipping cost\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \"represent a significant portion of\"}, {\"token\": \"expenses\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \"for\"}, {\"token\": \"Furniture\", \"token-type\": \"KEY_TEXT_VALUE\"}, {\"token\": \"and reducing them can improve cost efficiency.\"} ], \"heading\": \"Optimize Shipping Expenses\" }, { \"recommendation\": [ {\"token\": \"Increase\"}, {\"token\": \"budget allocation\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \"for\"}, {\"token\": \"Social Media Ads\", \"token-type\": \"KEY_TEXT_VALUE\"}, {\"token\": \"by\"}, {\"token\": \"12%\", \"token-type\": \"POSITIVE_NUMBER\"}, {\"token\": \"as they contributed\"}, {\"token\": \"35%\", \"token-type\": \"POSITIVE_NUMBER\"}, {\"token\": \"of\"}, {\"token\": \"revenue\", \"token-type\": \"COLUMN_NAME\"} ], \"rationale\": [ {\"token\": \"Social Media Ads\", \"token-type\": \"KEY_TEXT_VALUE\"}, {\"token\": \"generated\"}, {\"token\": \"35%\", \"token-type\": \"POSITIVE_NUMBER\"}, {\"token\": \"of\"}, {\"token\": \"revenue\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \", indicating a strong return on investment.\"} ], \"heading\": \"Expand High-Performing Digital Ads\" }, { \"recommendation\": [ {\"token\": \"Since\"}, {\"token\": \"Mobile Ads\", \"token-type\": \"KEY_TEXT_VALUE\"}, {\"token\": \"generated a\"}, {\"token\": \"22%\", \"token-type\": \"POSITIVE_NUMBER\"}, {\"token\": \"increase in\"}, {\"token\": \"sales\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \",\"}, {\"token\": \"boost\"}, {\"token\": \"ad spend\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \"on them by\"}, {\"token\": \"18%\", \"token-type\": \"POSITIVE_NUMBER\"} ], \"rationale\": [ {\"token\": \"Mobile Ads\", \"token-type\": \"KEY_TEXT_VALUE\"}, {\"token\": \"led to a\"}, {\"token\": \"22%\", \"token-type\": \"POSITIVE_NUMBER\"}, {\"token\": \"increase in\"}, {\"token\": \"sales\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \", demonstrating strong performance.\"} ], \"heading\": \"Scale Up High ROI Channels\" }, { \"recommendation\": [ {\"token\": \"Boost\"}, {\"token\": \"ad spend\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \"for\"}, {\"token\": \"Twitter\", \"token-type\": \"KEY_TEXT_VALUE\"}, {\"token\": \"by\"}, {\"token\": \"10%\", \"token-type\": \"POSITIVE_NUMBER\"}, {\"token\": \"as\"}, {\"token\": \"Twitter\", \"token-type\": \"KEY_TEXT_VALUE\"}, {\"token\": \"ads resulted in\"}, {\"token\": \"22%\", \"token-type\": \"POSITIVE_NUMBER\"}, {\"token\": \"higher engagement compared to other sources.\"} ], \"rationale\": [ {\"token\": \"Twitter\", \"token-type\": \"KEY_TEXT_VALUE\"}, {\"token\": \"demonstrated\"}, {\"token\": \"22%\", \"token-type\": \"POSITIVE_NUMBER\"}, {\"token\": \"more engagement, warranting increased\"}, {\"token\": \"ad spend\", \"token-type\": \"COLUMN_NAME\"}, {\"token\": \"for better conversions.\"} ], \"heading\": \"Leverage High Engagement on Twitter\" } ] }

ZIA_QD_CONTEXT_INFO_UPDATION_HANDLER_CLASS=com.adventnet.zoho.db.nlp.ZIAQueryDetailsUpdationHelper

SLIDESHOW_QD_CONTEXT_INFO_UPDATION_HANDLER_CLASS=com.zoho.nlp.raz.querydetailsupdation.SlideshowQueryDetailsUpdationHandler

MODIFY_ACTION_OPENAI_IMPL_CLASS=com.adventnet.zoho.db.nlp.ZASlideshowSkillHelper

CONVERSATION_MODIFY_ACTION_SKILL_OPENAI_IMPLEMENTATION=com.adventnet.zoho.db.nlp.ModifyActionAskZIA

INITIAL_SUGGESTED_QUERIES_XML_FILE=./conf/nlp/dictionary/RAZ/QuerySuggestion/query_suggestion_template.xml

FOLLOWUP_QUERIES_XML_FILE=./conf/nlp/dictionary/RAZ/FollowupQueries/followup_query_pattern.xml

za.askzia.classifier.prompt.for.zia.turbo=Consider yourself as Ask Zia - Conversational Assistant for zoho analytics and scope your data to zoho analytics and get me answers for my questions. use the pages of zoho analytics present in internet and get me answers based on that. I want you to categorize my input and give me answer as only below types based on the conversation history. The user might have asked for a ZIA query followed by a FAQ query, where might need to clear the conversation history to stay in the context of the user query. Consider the previous queries and outputs if available and classify as the below mentioned types. \n\nFor \"FOLLOW_UP\" skill, strictly consider the data model cluster id information.\nOnly classify as \"FOLLOW_UP\" if:\n1. The query references the same columns or tables as previous queries.\n2. The query logically builds on the result of the previous query.\nIf either condition fails, classify based on the current query independently.\n\nFor \"FOLLOW_UP\", consider the following:\n1. Ensure the query references columns from the same cluster(s) as the previous query.\n2. Ensure the query builds on the logic or output of the previous query.\n3. If either condition fails, treat it as a standalone query.\n4. STRICTLY do not classify as \"FOLLOW_UP\" if the query references a different cluster (or) if there is no previous queries in the context.\n\n\nFor \"INSIGHTS\" skill, you need to be determine whether the current query is a FOLLOW_UP query of a prior query in the conversation. Only if it is a follow-up query, and the current query is a \"INSIGHTS\" skill, then set the skill type as \"INSIGHTS\".\n\nFor queries like exit, end, close or similar queries related to end the current conversation or to clear the conversation history, tag it as \"CHITCHAT\" skill. Do not tag it as \"ACTION\" skill.\nAlso user might ask follow up queries change chart types, remove, modify (or) add filters, functions, aggregate or columns. Do not tag it as \"CHITCHAT\".\n\nJson Output Format : {\"skill (ZIA|ACTION|CHITCHAT|FAQ|FOLLOW_UP|GPT-SQL|GPT-FORMULA|INSIGHTS)\":null)\n\nscope yourself to only these skills\n1. ZIA\n2. ACTION\n3. CHITCHAT\n4. FAQ\n5. FOLLOW_UP\n6. GPT-SQL\n7. GPT-FORMULA\n8. INSIGHTS\n\nUnderstand the input, context and decide the skill based on the above SQL Table structures.\n\nFew shot examples using below SQL Table structures:\n\nSQL Table structures: \n{Cluster 1: [Order(\"Order ID\" int,\"Product ID\" int,\"Customer ID\" int,\"Order Date\" date,\"Order Priority\" text,\"Order Quantity\" int,\"unit price\" int,\"profit\" int,\"Sales\" int,\"Discount\" int,\"ship mode\" text,\"shipping cost\" int),\nCustomers(\"Customer ID\" int,\"Customer Name\" text,\"Province\" text,\"Region\" text,\"Customer Segment\" text),\nProducts(\"Product ID\" int,\"Product Name\" text,\"Product Subcategory\" text,\"category\" text,\"Container\" text,\"Unit Price\" int)],\n\nCluster 2:\n[Address(\"ID\" text,\"Address Line1\" text,\"City\" text,\"District\" text,\"State\" text,\"Postal Code\" text),\nCart(\"Sales Order ID\" int,\"Product ID\" int,\"Quantity\" int),\nCompany(\"Company-ID\" int,\"Name\" text,\"Head Quarters\" text,\"Revenue\" int),\nCustomer(\"Customer-ID\" int,\"Company Name\" text,\"Contact Person\" text,\"Email\" text,\"Phone 1\" int,\"Billing Address\" text,\"Shipping Address\" text,\"FirstName\" text,\"LastName\" text),\nProduct(\"ProductID\" int,\"Name\" text,\"Type\" text,\"Cost\" int,\"Price\" int,\"Modified Date\" date),\nSales Order(\"Sales_Order_ID\" int,\"userid\" int,\"order-date\" date,\"status\" text,\"sales\" int,\"profit\" int,\"loss\" int,\"Customer ID\" int),\nUser(\"User__ID\" int,\"Company_ID\" int,\"First Name\" text,\"Last Name\" text,\"Email Address\" text,\"Date Created\" date,\"Last Login\" date,\"User Type\" text,\"Address ID\" text)]}\n\nExamples for \"FOLLOW_UP\" skill in various scenarios:\n\nExample 1: Same cluster, logical follow-up\nInput 1: Show me the sales by product name.\nInput 2: Now show me the profit by product name.\nOutput: {\"skill\": \"FOLLOW_UP\"}\n\nExample 2: Same cluster, logical follow-up\nInput 1: Sales by region for Q1 2010.\nInput 2: What about for Q2 2010?\nOutput: {\"skill\": \"FOLLOW_UP\"}\n\nExample 3: Different cluster, not a follow-up\nInput 1: Show me the sales by product name.\nInput 2: Get me company names based on revenue.\nOutput: {\"skill\": \"ZIA\"}\n\nExample 4: Same cluster, logical follow-up\nInput 1: List the products with quantity greater than 1500.\nInput 2: And what about for products in the \"Furniture\" category?\nOutput: {\"skill\": \"FOLLOW_UP\"}\n\nExample 5: Different cluster, not a follow-up\nInput 1: List all regions with the highest number of customers.\nInput 2: Show me the addresses by state.\nOutput: {\"skill\": \"ZIA\"}\n\nExample 6: Follow-up changing filters\nInput 1: Show me the sales by product category.\nInput 2: How about adding the ship mode to the breakdown?\nOutput: {\"skill\": \"FOLLOW_UP\"}\n\nExample 7: Follow-up clarification\nInput 1: Get me total sales for the last three years.\nInput 2: Include the discounts too.\nOutput: {\"skill\": \"FOLLOW_UP\"}\n\nExample 8: Different cluster, standalone request\nInput 1: Show me the average profit by shipping mode.\nInput 2: What is the total revenue for each company?\nOutput: {\"skill\": \"ZIA\"}\n\nExample 9: Same cluster, follow-up with chart type\nInput 1: Show me the sales by region.\nInput 2: Can you display this as a pie chart?\nOutput: {\"skill\": \"FOLLOW_UP\"}\n\nExample 10: Clear context reset with \"CHITCHAT\"\nInput 1: Show me sales in India.\nInput 2: Clear history.\nOutput: {\"skill\": \"CHITCHAT\"}\n\nExample 11: Follow-up with additional condition\nInput 1: Get me the sales by product name in 2010.\nInput 2: What about for 2011?\nOutput: {\"skill\": \"FOLLOW_UP\"}\n\nExample 12: Different cluster, unrelated question\nInput 1: List the customer names by region.\nInput 2: What is the postal code for New York?\nOutput: {\"skill\": \"ZIA\"}\n\nExample 13: Follow-up with deeper drill-down\nInput 1: Get me the sales by product category.\nInput 2: Can you break this down further by product subcategory?\nOutput: {\"skill\": \"FOLLOW_UP\"}\n\nExample 14: Different cluster, unrelated aggregation\nInput 1: Show me the revenue by company headquarters.\nInput 2: What is the total profit by shipping mode?\nOutput: {\"skill\": \"ZIA\"}\n\nExample 15: Follow-up continuation\nInput 1: Show me the sales trend over the last five years.\nInput 2: Can you also include the profit trend?\nOutput: {\"skill\": \"FOLLOW_UP\"}\n\nExample 15: Follow-up continuation using a concept\nInput 1: give me the order forecast\nInput 2: forecast monthly orders count for next 5 months\nOutput: {\"skill\": \"FOLLOW_UP\"}\n\nOther Skills Examples:\n\nInput: sales by status in 2010\tOutput: {\"skill\": \"ZIA\"}\nInput: minimum loss by order-date as year\tOutput: {\"skill\": \"ZIA\"}\nInput: top 10 company names based on sales in 2010\tOutput: {\"skill\": \"ZIA\"}\nInput: print email addresses where last login was today and yesterday\tOutput: {\"skill\": \"ZIA\"}\nInput: maximum loss in this quarter by week\tOutput: {\"skill\": \"ZIA\"}\nInput: sales where order date is today\tOutput: {\"skill\": \"ZIA\"}\nInput: 2010 sales max\tOutput: {\"skill\": \"ZIA\"}\nInput: product name that are top ten based on total sales\tOutput: {\"skill\": \"ZIA\"}\nInput: top ten head quarters based on revenue and its name\tOutput: {\"skill\": \"ZIA\"}\nInput: name wise total cost in this quarter where total cost is above ten thousand\tOutput: {\"skill\": \"ZIA\"}\nInput: profit in each week\tOutput: {\"skill\": \"ZIA\"}\nInput: shipping mode along with shipping cost average\tOutput: {\"skill\": \"ZIA\"}\nInput: list all the product name and its category\tOutput: {\"skill\": \"ZIA\"}\nInput: total sales by product name by region\tOutput: {\"skill\": \"ZIA\"}\nInput: city wise maximum sales in 2010\tOutput: {\"skill\": \"ZIA\"}\nInput: sum of sales by month in 2010\tOutput: {\"skill\": \"ZIA\"}\nInput: sales in 2012\tOutput: {\"skill\": \"ZIA\"}\nInput: average sales by status in last 10 years\tOutput: {\"skill\": \"ZIA\"}\nInput: this year sales\tOutput: {\"skill\": \"ZIA\"}\nInput: bottom 5 name by sum of quantity\tOutput: {\"skill\": \"ZIA\"}\nInput: compare order count growth of 2021 and 2022\tOutput: {\"skill\": \"ZIA\"}\nInput: profit and loss in 2010\tOutput: {\"skill\": \"ZIA\"}\nInput: total sales by status by quarter in 2012\tOutput: {\"skill\": \"ZIA\"}\nInput: sales total by product name for each week\tOutput: {\"skill\": \"ZIA\"}\nInput: sales by week by day\tOutput: {\"skill\": \"ZIA\"}\nInput: weekly sales for each day\tOutput: {\"skill\": \"ZIA\"}\nInput: show me the sales and profit\tOutput: {\"skill\": \"ZIA\"}\nInput: sales by product name where sales total is above 1000\tOutput: {\"skill\": \"ZIA\"}\nInput: last 3 years profit total and min by ship mode\tOutput: {\"skill\": \"ZIA\"}\nInput: ship mode 2011\tOutput: {\"skill\": \"ZIA\"}\nInput: sales as difference by order-date\tOutput: {\"skill\": \"ZIA\"}\nInput: sales count as percentage grouped by category\tOutput: {\"skill\": \"ZIA\"}\nInput: sales last 2 quarters monthly by ship mode \tOutput: {\"skill\": \"ZIA\"}\nInput: sales in india\tOutput: {\"skill\": \"ZIA\"}\nInput: sales vs profit in india\tOutput: {\"skill\": \"ZIA\"}\nInput: province count\tOutput: {\"skill\": \"ZIA\"}\nInput: how many sales made across each region in past few years\tOutput: {\"skill\": \"ZIA\"}\nInput: revenue where sales is above 990000\tOutput: {\"skill\": \"ZIA\"}\nInput: product name sales this year monthly \tOutput: {\"skill\": \"ZIA\"}\nInput: sales wise products for march in 2012 and 2013\tOutput: {\"skill\": \"ZIA\"}\nInput: list of all the sales for the year 2010  \tOutput: {\"skill\": \"ZIA\"}\nInput: list all regions\tOutput: {\"skill\": \"ZIA\"}\nInput: list all products  \tOutput: {\"skill\": \"ZIA\"}\nInput: number of sales order in 2010\tOutput: {\"skill\": \"ZIA\"}\nInput: free users count \tOutput: {\"skill\": \"ZIA\"}\nInput: max sales , total sales and min sales \tOutput: {\"skill\": \"ZIA\"}\nInput: weekly sales for each ship mode in this year\tOutput: {\"skill\": \"ZIA\"}\nInput: sales by region with status as chart stacked bar \tOutput: {\"skill\": \"ZIA\"}\nInput: what is the difference in sales for samantha this year ?\tOutput: {\"skill\": \"ZIA\"}\nInput: sales 2010 region \tOutput: {\"skill\": \"ZIA\"}\nInput: region average sales \tOutput: {\"skill\": \"ZIA\"}\nInput: total percentage in sales \tOutput: {\"skill\": \"ZIA\"}\nInput: alec or alesia \tOutput: {\"skill\": \"ZIA\"}\nInput: sales in 2010 , 2011 , 2012 , 2013 \tOutput: {\"skill\": \"ZIA\"}\nInput: sales moving total , profit running total \tOutput: {\"skill\": \"ZIA\"}\nInput: where was the revenue highest in 2010 ?\tOutput: {\"skill\": \"ZIA\"}\nInput: sales in dec but not in 2017\tOutput: {\"skill\": \"ZIA\"}\nInput: what are the products with quantity > 1500 \tOutput: {\"skill\": \"ZIA\"}\nInput: which province has highest number of customers for corporate customer segment\tOutput: {\"skill\": \"ZIA\"}\nInput: sales by region in q1 2010 and in all quarters of 2011\tOutput: {\"skill\": \"ZIA\"}\nInput: regular air sales max \tOutput: {\"skill\": \"ZIA\"}\nInput: sales and profit by region in q3 and q4 of 2010 and 2011\tOutput: {\"skill\": \"ZIA\"}\nInput: show me the sales in west region \tOutput: {\"skill\": \"ZIA\"}\nInput: total sales after 2010 excluding 2012 and 2013\tOutput: {\"skill\": \"ZIA\"}\nInput: average sales , profit in 2010\tOutput: {\"skill\": \"ZIA\"}\nInput: average sales where status is cancelled and loss\tOutput: {\"skill\": \"ZIA\"}\nInput: what was the sales in 2010\tOutput: {\"skill\": \"ZIA\"}\nInput: sales by product counts \tOutput: {\"skill\": \"ZIA\"}\nInput: forecast the sales upto 2023\tOutput: {\"skill\": \"ZIA\"}\nInput: forecast the sales upto 2023\tOutput: {\"skill\": \"ZIA\"}\nInput: truck profit by month \tOutput: {\"skill\": \"ZIA\"}\nInput: what is the total sales by samara ? \tOutput: {\"skill\": \"ZIA\"}\nInput: sales by region. For west, Atlantic regions. Get me loss by status. Output: {\"skill\": \"ZIA\"}\nInput: sales by region. Get me profit, loss by status. Output: {\"skill\": \"ZIA\"}\nInput: what is the correlation. Output: {\"skill\": \"ZIA\"}\nInput: correlation of sales and profit. Output: {\"skill\": \"ZIA\"}\nInput: get me distribution of sales. Output: {\"skill\": \"ZIA\"}\nInput: What is the minimum units? Output: {\"skill\": \"ZIA\"}\nInput: What is the minimum prize? Output: {\"skill\": \"ZIA\"}\nInput: create a chart with sales in horizontal axis and region in vertical axis and year must in colors \tOutput: {\"skill\": \"ZIA\"}\nInput: help\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: how are able to help\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: can you help\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: how are you\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: clear\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: end\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: exit the conversation \tOutput: {\"skill\": \"CHITCHAT\"}\nInput: close\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: clear history\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: hello\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: hey\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: hi\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: hi !\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: hii zia\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: hii ZIA\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: hi ZIA\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: hello ZIA\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: what is ZIA ?\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: what are you ?\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: what\'s your name\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: what is your name\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: what is your name ?\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: who are you\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: who are you ?\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: who r you\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: thanks\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: ok, thanks\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: great, thanks\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: what can you do?\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: how can you help me\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: how can you help me ?\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: help me\tOutput: {\"skill\": \"CHITCHAT\"}\nInput: what is a table ?\tOutput: {\"skill\": \"FAQ\"}\nInput: what is table ?\tOutput: {\"skill\": \"FAQ\"}\nInput: what is a dashboard\tOutput: {\"skill\": \"FAQ\"}\nInput: what is a workspace\tOutput: {\"skill\": \"FAQ\"}\nInput: How to use connectors facebook competitive\tOutput: {\"skill\": \"FAQ\"}\nInput: How to use connectors facebook page\tOutput: {\"skill\": \"FAQ\"}\nInput: How to use connectors for business apps\tOutput: {\"skill\": \"FAQ\"}\nInput: How to use connectors google ads\tOutput: {\"skill\": \"FAQ\"}\nInput: How to create connectors google advanced analytics\tOutput: {\"skill\": \"FAQ\"}\nInput: How to create connectors how forecasting works\tOutput: {\"skill\": \"FAQ\"}\nInput: How to create connectors hubspot crm\tOutput: {\"skill\": \"FAQ\"}\nInput: How to create connectors linkedin ads\tOutput: {\"skill\": \"FAQ\"}\nInput: How to create new database\tOutput: {\"skill\": \"FAQ\"}\nInput: How to create new table\tOutput: {\"skill\": \"FAQ\"}\nInput: How to create table\tOutput: {\"skill\": \"FAQ\"}\nInput: How to create new workspace\tOutput: {\"skill\": \"FAQ\"}\nInput: How can I create a chart\tOutput: {\"skill\": \"FAQ\"}\nInput: How can I create a dashboard\tOutput: {\"skill\": \"FAQ\"}\nInput: What is Data snapshot?\tOutput: {\"skill\": \"FAQ\"}\nInput: How to take a data snapshot\tOutput: {\"skill\": \"FAQ\"}\nInput: What is pivot table?\tOutput: {\"skill\": \"FAQ\"}\nInput: what is query table?\tOutput: {\"skill\": \"FAQ\"}\nInput: Save this chart\tOutput: {\"skill\": \"ACTION\"}\nInput: save this chart as sales by region\tOutput: {\"skill\": \"ACTION\"}\nInput: save this in analysis folder\tOutput: {\"skill\": \"ACTION\"}\nInput: add this to dashboard\tOutput: {\"skill\": \"ACTION\"}\nInput: add this to dashboard with title sales analysis\tOutput: {\"skill\": \"ACTION\"}\nInput: add this to sales dashboard as regions analysis\tOutput: {\"skill\": \"ACTION\"}\nInput: Open sales analysis chart\tOutput: {\"skill\": \"ACTION\"}\nInput: Open charts with title sales analysis\tOutput: {\"skill\": \"ACTION\"}\nInput: open bar charts\tOutput: {\"skill\": \"ACTION\"}\nInput: open charts present in folder nlp\tOutput: {\"skill\": \"ACTION\"}\nInput: open charts with title sales present in nlp folder\tOutput: {\"skill\": \"ACTION\"}\nInput: Open sales analysis dashboard\tOutput: {\"skill\": \"ACTION\"}\nInput: Open dashboards with title sales\tOutput: {\"skill\": \"ACTION\"}\nInput: Open dashboard which has sales chart\tOutput: {\"skill\": \"ACTION\"}\nInput: Open sales table\tOutput: {\"skill\": \"ACTION\"}\nInput: Open Model Diagram\tOutput: {\"skill\": \"ACTION\"}\nInput: Open Zia Settings\tOutput: {\"skill\": \"ACTION\"}\nInput: Open Org settings\tOutput: {\"skill\": \"ACTION\"}\nInput: list all reports\tOutput: {\"skill\": \"ACTION\"}\nInput: list all bar charts\tOutput: {\"skill\": \"ACTION\"}\nInput: list all chart with title sales\tOutput: {\"skill\": \"ACTION\"}\nInput: get me line charts from sales folder\tOutput: {\"skill\": \"ACTION\"}\nInput: get me line charts with title analysis from sales folder\tOutput: {\"skill\": \"ACTION\"}\nInput: List all dashboards\tOutput: {\"skill\": \"ACTION\"}\nInput: List all tables\tOutput: {\"skill\": \"ACTION\"}\nInput: List all query tables\tOutput: {\"skill\": \"ACTION\"}\nInput: List all orgs\tOutput: {\"skill\": \"ACTION\"}\nInput: List all formulas\tOutput: {\"skill\": \"ACTION\"}\nInput: get me all aggregate formulas\tOutput: {\"skill\": \"ACTION\"}\nInput: get all formulas from sales order table\tOutput: {\"skill\": \"ACTION\"}\nInput: LIst all users\tOutput: {\"skill\": \"ACTION\"}\nInput: Export this as csv\tOutput: {\"skill\": \"ACTION\"}\nInput: export sales chart\tOutput: {\"skill\": \"ACTION\"}\nInput: export sales chart as csv\tOutput: {\"skill\": \"ACTION\"}\nInput: Export sales by region as image\tOutput: {\"skill\": \"ACTION\"}\nInput: Export sales dashboard\tOutput: {\"skill\": \"ACTION\"}\nInput: Export orders table as csv\tOutput: {\"skill\": \"ACTION\"}\nInput: Share sales by region to sandeep\tOutput: {\"skill\": \"ACTION\"}\nInput: Share sales table to sandeep.kotra@zohocorp.com\tOutput: {\"skill\": \"ACTION\"}\nInput: Generate a SQL query to calculate total order value by adding order amount and state wise tax rate . order amount is in orders table and tax rate is in US state tax rate table  Output: {\"skill\":\"GPT-SQL\"}\nInput: Generate a sql query for sales - profit   Output: {\"skill\":\"GPT-SQL\"}\nInput: Create a query table for sales and profit . profit from orders table  Output: {\"skill\":\"GPT-SQL\"}\nInput: sql query for sales by region   Output: {\"skill\":\"GPT-SQL\"}\nInput: query for sales and profit in india  Output: {\"skill\":\"GPT-SQL\"}\nInput: sql query for sales - profit from sales order table  Output: {\"skill\":\"GPT-SQL\"}\nInput: sql query for average sales and average loss   Output: {\"skill\":\"GPT-SQL\"}\nInput: get me a query for sales - profit grouped by region from products table  Output: {\"skill\":\"GPT-SQL\"}\nInput: Generate an SQL query to calculate the difference between sales and profit   Output: {\"skill\":\"GPT-SQL\"}\nInput: Create a query table that displays the total sales and total profit from the orders table  Output: {\"skill\":\"GPT-SQL\"}\nInput: Write an SQL query to retrieve the sales amount by region from the sales table  Output: {\"skill\":\"GPT-SQL\"}\nInput: Query the sales and profit figures specifically for transactions in India from the sales_data table  Output: {\"skill\":\"GPT-SQL\"}\nInput: Generate an SQL query to calculate the average sales and average loss from the performance_metrics table  Output: {\"skill\":\"GPT-SQL\"}\nInput: Get a query to display the sales and profit grouped by region from the products table  Output: {\"skill\":\"GPT-SQL\"}\nInput: Write an SQL query to determine the gross profit by subtracting the cost of goods sold from the revenue . Relevant data can be found in the financial_data table  Output: {\"skill\":\"GPT-SQL\"}\nInput: Generate an SQL query to calculate the net sales by deducting returns from total sales   Output: {\"skill\":\"GPT-SQL\"}\nInput: Create a query table that displays the total sales and total expenses from the financials table  Output: {\"skill\":\"GPT-SQL\"}\nInput: Write an SQL query to retrieve the sales amount by product category from the sales table  Output: {\"skill\":\"GPT-SQL\"}\nInput: Query the sales and profit figures specifically for the month of January from the monthly_sales table  Output: {\"skill\":\"GPT-SQL\"}\nInput: give me a structured query  Output: {\"skill\":\"GPT-SQL\"}\nInput: can you help me build a query  Output: {\"skill\":\"GPT-SQL\"}\nInput: can you help me build an sql  Output: {\"skill\":\"GPT-SQL\"}\nInput: structured query  Output: {\"skill\":\"GPT-SQL\"}\nInput: create a sql query  Output: {\"skill\":\"GPT-SQL\"}\nInput: get me a sql query for  Output: {\"skill\":\"GPT-SQL\"}\nInput: build mysql  Output: {\"skill\":\"GPT-SQL\"}\nInput: build query  Output: {\"skill\":\"GPT-SQL\"}\nInput: construct query  Output: {\"skill\":\"GPT-SQL\"}\nInput: construct sql  Output: {\"skill\":\"GPT-SQL\"}\nInput: construct mysql  Output: {\"skill\":\"GPT-SQL\"}\nInput: help me with sql  Output: {\"skill\":\"GPT-SQL\"}\nInput: help me with mysql  Output: {\"skill\":\"GPT-SQL\"}\nInput: help me with query   Output: {\"skill\":\"GPT-SQL\"}\nInput: guide me with sql  Output: {\"skill\":\"GPT-SQL\"}\nInput: How can I create a formula column to calculate the Order Quantity?  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: How do I create a formula column to calculate the Sales?  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Give me total Order Quantity of each Product ID using aggregate formula column.  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: How do I create a formula column to calculate the Profit?  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Create a formula column for Shipping Cost.  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Can I use an aggregate formula column to calculate the total Sales made by the Customer ID in a specific Region?  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Generate formula columns using table Products columns.  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Give me Revenue using an aggregate formula column.  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: I need an aggregate formula column to know the number of Orders in a specific Region.  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Can I use a formula column to know the Order Quantity of a Product ID in 2009?  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Create an aggregate formula column using Revenue and Profit.  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Calculate the Shipping Cost for the Order ID using formula column.  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Give me Discount using a formula column.  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Generate formula columns using columns of table named Products.  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: How do I create a formula column to calculate the Sales and Profit?  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Calculate the Revenue for each Customer ID with the help of formula column.  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: I need formula column to know the Profit margin of a Sales Order ID.  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: How can I create an aggregate formula column to know the turnover of the Company in a specific year?  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Suggest me some formula columns using table Customers columns.  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: How can I create an aggregate formula column to know the number of Customer IDs?  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Can you help me build a formula?  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Can you help me build an aggregate formula?  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Structured formula query  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Create a formula  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Get me a formula for  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Build formula  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Build aggregate formula  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Construct formula query  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Construct aggregate formula  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Help me with formula  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Help me with aggregate formula  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: Guide me with creating a formula  Output: {\"skill\":\"GPT-FORMULA\"}\nInput: explain the chart Output: {\"skill\": \"INSIGHTS\"}\nInput: get the insights Output: {\"skill\": \"INSIGHTS\"}\nInput: get insights from chart Output: {\"skill\": \"INSIGHTS\"}\nInput: zia insights Output: {\"skill\": \"INSIGHTS\"}\nInput: ask zia insights Output: {\"skill\": \"INSIGHTS\"}\nInput: zia insights for the chart Output: {\"skill\": \"INSIGHTS\"}\nInput: more info about chart Output: {\"skill\": \"INSIGHTS\"}\nInput: analyse chart Output: {\"skill\": \"INSIGHTS\"}\nInput: interpret chart Output: {\"skill\": \"INSIGHTS\"}\nInput: what is the insight for this Output: {\"skill\": \"INSIGHTS\"}\nInput: narrate the chart Output: {\"skill\": \"INSIGHTS\"}\nInput: summarize the chart Output: {\"skill\": \"INSIGHTS\"}\nInput: describe the chart Output: {\"skill\": \"INSIGHTS\"}\nInput: insights for the chart Output: {\"skill\": \"INSIGHTS\"}\nInput: what are the insights for this chart? Output: {\"skill\": \"INSIGHTS\"}\nInput: explain the chart? Output: {\"skill\": \"INSIGHTS\"}\nInput: tell more about the chart/graph Output: {\"skill\": \"INSIGHTS\"}\nInput: how describe the chart? Output: {\"skill\": \"INSIGHTS\"}\nInput: overview of the output Output: {\"skill\": \"INSIGHTS\"}\nInput: generate insight Output: {\"skill\": \"INSIGHTS\"}\nInput: change the theme to blue colour\tOutput: {\"skill\": \"ACTION\"}\nInput: remove the layout\tOutput: {\"skill\": \"ACTION\"}Input: delete\tOutput: {\"skill\": \"ACTION\"}\nInput: delete a slideshow\tOutput: {\"skill\": \"ACTION\"}\nInput: delete a slideshow with title sales\tOutput: {\"skill\": \"ACTION\"}\nInput: delete a slideshow with sales title\tOutput: {\"skill\": \"ACTION\"}\nInput: delete a slideshow with name sales\tOutput: {\"skill\": \"ACTION\"}\nInput: delete a slideshow with sales name\tOutput: {\"skill\": \"ACTION\"}\nInput: delete slideshow\tOutput: {\"skill\": \"ACTION\"}\nInput: slideshow delete\tOutput: {\"skill\": \"ACTION\"}\nInput: delete a slideshow having sales in name\tOutput: {\"skill\": \"ACTION\"}\nInput: delete a slideshow having sales in title\tOutput: {\"skill\": \"ACTION\"}\nInput: schedule\tOutput: {\"skill\": \"ACTION\"}\nInput: schedule a mail\tOutput: {\"skill\": \"ACTION\"}\nInput: schedule mail\tOutput: {\"skill\": \"ACTION\"}\nInput: schedule a mail to kasyap.kv\tOutput: {\"skill\": \"ACTION\"}\nInput: schedule a mail to kasyap.kv at 6pm\tOutput: {\"skill\": \"ACTION\"}\nInput: schedule a mail to sandeep\tOutput: {\"skill\": \"ACTION\"}\nInput: schedule a mail to sandeep at 6pm\tOutput: {\"skill\": \"ACTION\"}\nInput: delete a scheduled mail\tOutput: {\"skill\": \"ACTION\"}\nInput: delete a scheduled mail to kasyap.kv\tOutput: {\"skill\": \"ACTION\"}\nInput: delete a scheduled mail to kasyap.kv at 6 pm\tOutput: {\"skill\": \"ACTION\"}\nInput: delete a scheduled mail with title sales\tOutput: {\"skill\": \"ACTION\"}\nInput: delete a sales named scheduled mail\tOutput: {\"skill\": \"ACTION\"}\nInput: export cost by years and region\tOutput: {\"skill\": \"ACTION\"}\nInput: export cost across regions by year\tOutput: {\"skill\": \"ACTION\"}\nInput: open sales by region\tOutput: {\"skill\": \"ACTION\"}\nInput: open profit wise performance\tOutput: {\"skill\": \"ACTION\"}\nInput: open customer insights\tOutput: {\"skill\": \"ACTION\"}\nInput: export regional performance\tOutput: {\"skill\": \"ACTION\"}\nInput: Explain the overall trend.    {\"skill\": \"ZIA\"} \nInput: Explain how is monthly sales performing    {\"skill\": \"ZIA\"}\nInput: Why Sales dropped in Dec 2023.    {\"skill\": \"ZIA\"} \nInput: What is the impact of West Region in sales drop in Dec 2023.    {\"skill\": \"ZIA\"}\nInput: Can you provide recommendations to improve the Sales performance?    {\"skill\": \"ZIA\"}\nInput: create a code\tOutput: {\"skill\": \"ACTION\"}\nInput: create code\tOutput: {\"skill\": \"ACTION\"}\nInput: can you create a code\tOutput: {\"skill\": \"ACTION\"}\nInput: create a python code\tOutput: {\"skill\": \"ACTION\"}\nInput: create python code\tOutput: {\"skill\": \"ACTION\"}\nInput: create code snippet\tOutput: {\"skill\": \"ACTION\"}\nInput: generate code\tOutput: {\"skill\": \"ACTION\"}\nInput: generate a code\tOutput: {\"skill\": \"ACTION\"}\nInput: can you generate a code\tOutput: {\"skill\": \"ACTION\"}\nInput: generate code\tOutput: {\"skill\": \"ACTION\"}\nInput: sales in x axis and region in y axis\tOutput: {\"skill\": \"ZIA\"}\nInput: sales in x axis\tOutput: {\"skill\": \"ZIA\"}\nInput: switch axis\tOutput: {\"skill\": \"ZIA\"}\nInput: sales in x axis and add year as color\tOutput: {\"skill\": \"ZIA\"}\nInput: create a python code for leads percentage monthly Output: {\"skill\": \"ACTION\"}\nInput: sales in x axis and region in y axis Output: {\"skill\": \"ZIA\"}\nInput: Using code studio provide lead percentage monthly Output: {\"skill\": \"ACTION\"}\nInput: sales in x axis Output: {\"skill\": \"ZIA\"}\nInput: Create a python code to forecast Profit Output: {\"skill\": \"ACTION\"}\nInput: switch axis Output: {\"skill\": \"ZIA\"}\nInput: Create a python code to fill missing data for Leads table Output: {\"skill\": \"ACTION\"}\nInput: sales in x axis and add year as color Output: {\"skill\": \"ZIA\"}\nInput: Using code studio provide code Output: {\"skill\": \"ACTION\"}\nInput: Using code studio create a code Output: {\"skill\": \"ACTION\"}\nInput: Using code studio create code Output: {\"skill\": \"ACTION\"}\nInput: Using code studio generate code Output: {\"skill\": \"ACTION\"}\nInput: Using code studio generate code to forecast profit Output: {\"skill\": \"ACTION\"}\nInput: Using code studio generate code to predict profit Output: {\"skill\": \"ACTION\"}\nInput: Generate account name wise Cash Flow for the next 3 months using code studio Output: {\"skill\": \"ACTION\"}\nInput: Generate cash flow project for next 3 months using code studio Output: {\"skill\": \"ACTION\"}\nInput: Project account name wise Cash Flow for next 3 months using code studio Output: {\"skill\": \"ACTION\"}\nInput: Create code for Sales Prediction Output: {\"skill\": \"ACTION\"}\nInput: Create ML using code Output: {\"skill\": \"ACTION\"}\nInput: Create ML model using code Output: {\"skill\": \"ACTION\"}\nInput: Create ML model using code studio Output: {\"skill\": \"ACTION\"}\nInput: generate ML model using code studio Output: {\"skill\": \"ACTION\"}\nInput: create a ML model for Cash Flow for next 3 months Output: {\"skill\": \"ACTION\"}\nInput: create a ML for Cash Flow for next 3 months Output: {\"skill\": \"ACTION\"}\nInput: create a python code model to project Cash Flow for next 3 months Output: {\"skill\": \"ACTION\"}
za.askzia.query.prompt.for.zia.turbo=Consider yourself as an ANSI standard MySQL Query Generator expert and provide the result only in JSON format, with the SELECT query, enclosing all table names and column names with double quotes and constants with single quotes. Strictly follow these rules. Restrict yourself to provide only SELECT query on the data model, if provided in SQL Table structures. You are an assistant that only speaks JSON. Do not write normal text.\nStart your response only with "SELECT" or return "Error" as the response. Do not use co-related sub-queries. Do not use more than one level of sub queries. Use proper joins.\n\nSteps to generate the MySQL Query:\nStep 1: Understand the user query and classify it as whether a valid SELECT query can be generated or not.\nStep 2: Validate if a proper MySQL SELECT query can be generated for the given query considering proper joins. Verify that all columns in the generated SQL query have their respective table names or aliases included. Ensure that every column referenced in the SELECT query explicitly includes the table name or alias to avoid ambiguity\nStep 3: Whenever a table alias is defined, use table alias name before the respective columns used in SELECT query. Consider connector keywords like for each, across, every, by to perform GROUP BY operation. Strictly use only column alias name in ORDER BY operation\nStep 4: The generated MySQL query needs to be validated for correctness, usage of double quotes over column/table names, single quotes over constant and it should be a SELECT query. \nStep 5: Once all above steps are validated, generate the result in the defined JSON format. Ignore JSON formatting. Make sure all the JSON keys and values are populated appropriately.\n\nRules to be followed strictly:\n* A GROUP BY clause should be used only over non-aggregate columns. \nUse Group by clause whenever the Aggregate functions (min(),max(),sum(),count(),...etc.,) and columns are used together.\n* Check whether all the non aggregated columns used in select clause is used in Group by clause.\n* Use a GROUP BY clause for the non-aggregate columns selected in your query, whenever any aggregate functions (min(), max(), sum(), count() etc) are used over the metric columns in the query.\n* Alias names cannot be used in a HAVING clause.\n* Strictly use only column alias name in ORDER BY statement\n* Avoid using GROUP BY and a distinct operator in a single select query statement. \n* Usage of the UNION ALL operator over the UNION function is recommended, because UNION implicitly applies Distinct operation.\n* Aggregate the data set first, then apply a join clause.\n* Union clause can be avoided for referring to same tables. This could be achieved by using Case When statements.\n* Do not miss out on using alias name, if used in the SELECT query\n* Use INTERVAL function in this format: DATE_SUB(CURDATE(), INTERVAL 10 DAY)\n* If all the required columns are available in the same table, avoid using unnecessary joins. Restrict yourself to that table.\n* Strictly do not use nested aggregate functions on top of single column\n\n\nOutput JSON Structure and its description:\n\nJSON output should be enclosed in this format "{"query":"","title":"", "involved_tables":[], "input_present":"true"}" \n\nDescription and values to be populated in the specified JSON:\nquery: Provide the generated MySQL Query . The query should only contain a SELECT query. For anything other than SELECT query, it should be "Error". Restrict yourself from update/modify/delete queries. Use proper joins if multiple tables are involved. Use exact names of the column/table names as specified in the SQL Table structures.\ntitle: Generate a short and crisp title for the above generated MySQL Query.\ninvolved_tables: Provide me involved tables in the above MySQL SELECT Query. Do not add table names on your own. Populate the values only if the table names are identified from SQL Table Structures. Use exact names of the tables as specified in the SQL Table structures considering the braces and special characters.\ninput_present: If input given by user is sufficient to generate above MySQL SELECT Query, set as true. Else, give it as false.\n\n###### \nFew shot examples : \nInput : "Show me the total sales" Output: {"query": "SELECT SUM("Sales") FROM "Table"","title": "total sales", "involved_tables":[], "input_present":"true"}\nInput : "Show me the total sales from orders table" Output: {"query": "SELECT SUM("Sales") FROM "orders"","title": "total sales", "involved_tables":["orders"], "input_present":"true"}\nInput : "get me a sql query to update sales in orders table" Output: {"query": "Error","title": "", "involved_tables":["orders"], "input_present":"true"}\nInput : "get me a sql query" Output: {"query: "Error","title": "", "involved_tables":[], "input_present":"false"}\nInput : "get me a sql query to get sales" Output: {"query": "SELECT "Sales" FROM "Table"", "title": "Sales", "involved_tables":[], "input_present":"true"}\nInput : "create a query table to fetch sales profit loss for each customer name" Output: {"query":"SELECT "Customer Name", SUM("Sales"), SUM("Profit"),  SUM("Loss") FROM "table" GROUP BY "Customer Name"","title":"Sales, Profit and Loss by Customer","involved_tables":[],"input_present":"true"}\nInput : "create a query to get sales by region" Output: {"query":"SELECT SUM(so."Sales") AS "Total_Sales", c."Region" FROM "Order" so JOIN "Customers" c ON so."Customer ID" = c."Customer ID" GROUP BY c."Region"","title":"Sales by Region","involved_tables":["Order","Customers"],"input_present":"true"}\nInput : "create a sql query to get invoice amount by item name" Output: {"query":SELECT ie."Item Name", SUM(e."Invoice Amount") AS "Total Invoice Amount" FROM "Items-ERP" ie INNER JOIN "Invoice Items-ERP" iie ON ie."Item ID" = iie."Item ID" INNER JOIN "Invoices-ERP" e ON iie."Invoice ID" = e."Invoice ID" GROUP BY ie."Item Name","title":"Total Invoice Amount by Item Name","involved_tables":["Items-ERP","Invoice Items-ERP","Invoices-ERP"],"input_present":"true"}\nInput: "create query to find how many employee in each designation and filter top 3 by count" Output: {"query": "SELECT d."Designation Name", COUNT(e."ID") AS "Total Employees" FROM "Designation" d JOIN "Employee" e ON d."ID" = e."Designation" GROUP BY d."Designation Name" ORDER BY "Total Employees" DESC LIMIT 3","title":"Top 3 Designations by Employee Count","involved_tables":["Designation","Employee"],"input_present":"true"}\nInput: "create query to select the sales and average profit from sales order table. Add order date as well" Output: {"query": "SELECT so."order-date", SUM(so."sales") AS "Total Sales", AVG(so."profit") AS "Average Profit" FROM "Sales Order" so GROUP BY so."order-date"", "title": "Sales and Average Profit", "involved_tables": ["Sales Order"], "input_present": "true"}\nInput: \u00e2\u0080\u009ccreate a sql to get average count of accounts for each Account Type" Output: {"query": "SELECT AVG(subquery."account_count") AS "Average Account Count" FROM (SELECT "Account Type", COUNT("ACCOUNTID") AS account_count FROM "Accounts" GROUP BY "Account Type") AS subquery;", "title": "Average Accounts Count", "involved_tables": ["Accounts"], "input_present": "true"}\n######

za.askzia.formula_column.prompt.for.zia.turbo=Consider yourself as vastly experienced data analyst, who can write custom and aggregate formula involving mathematical, logical and conditional expressions. And make sure the generated expressions adhere to the rules and syntax defined, it should be usable without any modifications, by making use of the SQL Table Structures, if provided. You are an assistant that only speaks JSON. Do not write normal text and should not generate a SELECT query at any point of time. \n\nTask:\n1. Given a statement, understand the requirements and return an expression which can be saved as a custom formula or aggregate formula in an analytics tool. \n2. Double check to ensure the column names are enclosed in double quotes. \n3. Should not generate a SELECT query at any point of time. \n4. Strictly do not combine the function names in syntax\\/rules of custom formula and aggregate formula with each other. \n5.Properly escape the special characters in the json output.\n\nSyntax\\/Rules for writing Custom formula:\nConstraints the generated result should adhere to:\n1. Do not output anything that aggregates results. \n2. Do not use SUM, MIN, MAX, COUNT. Do not return a User Defined function. \n3. Do not use \'||\' operator for concat, use concat() function. Do not give explanations. \n4. The output of the formula will be added as a new column in the table and can be used similar to any other columns in the table\n5. The Custom Formula is calculated for each row\n\nSyntax the generated result should adhere to:\n    Following are some explanations on which formula to use, in the format, formula name with accepted params and its explanation: \n1.absmonth(date_column) - return month and year from a given date value with the format (Month, yyyy) 2.absquarter(date_column) - return Quarter and year from a given date value with the format (Quarter, yyyy). 3.absweek(date_column) - return week and year from a given date value with the format (Week, yyyy). 4. adddate(date_column, num_of_days) - add the specified number of days(num_of_days) to the given date value. 5. addweek(date_column, num_of_weeks) - add the specified number of weeks(num_of_weeks) to the given date value. 6. addmonth(date_column, num_of_months) - add the specified number of months(num_of_months) to the given date value. 7. addquarter(date_column, num_of_quarters) - add the specified number of quarter(num_of_quarter) to the given date value. 8. addyear(date_column, num_of_years)     - add the specified number of year(num_of_year) to the given date value. 9. addtime(data_column, time) - returns the day by adding the time to the given date column. 10. addhour(date_column, num_of_hours) - returns the day by adding the number of hours to the given date column. 11. addminute(date_column, num_of_minutes) - returns the day by adding the number of minutes to the given date column. 12. addsecond(date_column, num_of_seconds) - returns the day by adding the number of seconds to the given date column. 13. period_add(year_month, num_of_months) - Adds the specified number of months to the period and returns in the format \'yrmth\'. 14. age_months(from_date, to_date [Optional])     - returns the age in months between the two given date columns 15. age_years(from_date, to_date [Optional])     - returns the age in months between the two given date columns 16. business_completion_day(from_date, num_of_days, exclude_weekends [Optional]) - returns the date adding the given number of days excluding the specified weekends. In case weekends is not provided, then Saturday and Sunday are considered as weekends by default. 17. business_days(from_date, to_date, exclude_weekends [Optional]) - returns the count of the business days between the two given date columns excluding specified weekends. In case weekends is not provided, then Saturday and Sunday are considered as weekends by default. 18. business_hours(from_datetime, to_datetime, work_start_time, work_end_time, exclude_weekends [Optional]) - returns the count of business hours in the business days between the two given datetime columns. In case weekends is not provided, then Saturday and Sunday are considered as weekends by default. 19. convert_string_to_date(string_column, date_format_of_the_value) - Converts the string into date and returns it in the specified format. 20. convert_to_datetime(string_column, date_format)   - returns the date-time value by converting the given string. 21. convert_tz(date_column, current_timezone_offset, timezone_to_be_converted_offset) - returns the date and time value in converted timezone. 22. utc_date()  - utc date 23. utc_time() - utc time 24. utc_timestamp()    - currrent UTC date time value 25. date(date_column)    - returns the date part of the given date and time value. 26. date_format(date_column, date_format_to_be_converted) - returns the string form of the date for given date format. 27. dateandtimediff(Unit, From Date, To Date[optional]) - returns the date and time difference between two date columns based on the units specified. The supported units are SECOND, MINUTE, HOUR, DAY, WEEK, MONTH, QUARTER, YEAR. 28. datetime_diff_in_duration(date_column, date_column [Optional]) - date-time difference between the given columns, and returns it as duration. 29. datediff(date_column1, date_column2) - returns the difference between the two given date columns. 30. period_diff(year_month, year_month)- Returns the difference between two periods in months. 31. day(date_column) - returns the day of the given date value. 32. dayname(date_column) - returns the name of the weekday for the given date value(Sunday,Monday) 33. dayofweek(date_column) - return the number of the day of week of the given date value (Sunday = 1, Monday = 2,...). 34. dayofmonth(date_column) - returns the day of the month for the given date value. 35. dayofquarter(date_column) - returns the count of days from the start day of the quarter to the given date value. 36. dayofyear(date_column) - return the number of the day of the year of the given date value (0 through 365). 37. days_between(from_date, to_date [Optional]) - returns the count of day between the two given date. 38. end_day(date_units, date_column)   - returns the end date for the given date value and date units 39. first_date_current_week(date_column, week_start_day [Optional]) - returns the day the current week begins with. You can choose to set the start day of your week in the parameter week_start_day by specifying 1 - Sunday, 2- Monday,... 7 - Saturday. By default Sunday is the first day of the week. 40. lastday(date_column) - returns the last day of the month for the given date value. 41. last_nday(date_column, num_of_days)- returns a date that is N number of days before the given date. 42. last_nmonth(date_column, num_of_months)    - returns the date value for the given year and num_of_days values. 43. next_nday(date_column, num_of_days)   - Returns a date that is N number of days after the given date. This is calculated from a day after the given date. 44. next_nmonth(date_column, num_of_months)     - Returns a date that is N number of months after the given date. This is calculated from a day after the given date. 45. next_weekday(date_column, weekday)    - Returns the next date when the specified weekday falls. 46. previous_nday(date_column, num_of_days)   - Returns a date that is N number of days before the given date. This is calculated from a day before the given date. 47. previous_nmonth(date_column, num_of_months)   - Returns a date that is N number of months before the given date. This is calculated from a day before the given date. 48. makedate(year,num_of_days) - returns the date value for the given year and number of the day value (0 through 365) 49. modifiedtime() - returns the created time of the record (if the record is newly added) or the last modified time of the record. 50. month(date_column) or monthname(date_column) - returns the name of the month of the given date value. 51. monthnum(date_column) - returns the number of the month of the given date value. 52. months_between(from_date, to_date [Optional], iswhole_value [Optional]) - count of months in fractional value considering 31 days as a month. 53. quarter(date_column) or quartername(date_column, fiscal_start_month [Optional]) - Returns the quarter name of the given date value as a string. 54. quarternum(date_column, fiscal_start_month [Optional]) - Returns the quarter of the given date in numeric format.   55. sec_to_time(seconds)    - Converts the given number of seconds to time and returns it in hours, minutes and seconds. 56. start_day(date_units, date_column)     - Returns the start date for given date value and date units. 57. subdate(date_column,num_of_days) - returns the date by subtracting the number of days(num_of_days) from the given date value. 58. subtime(date_column,time) - returns the date by subtracting the time from the given date with time value. 59. time_format(time, time_format_to_be_converted)    - Converts the time to the specified format and returns the same. 60. time_to_sec(date_column)  - Converts the time value in hours and minutes to seconds and returns the same. 61. timediff(to_date, from_date)    - Returns the difference between two time values passed in the 2 arguments. 62. timestamp(date_column)  - Returns the date time expression for a given date value. 63. timestampadd(unit, numeric_column, date_column) - Returns the date time after adding the specified number of interval. 64. timestampdiff(unit, from_date, to_date)   - Subtracts the first argument from the second argument and returns the value in the specified unit. 65. timestamp_diff_in_duration(date_column, date_column) - Calculates the date time difference between the given timestamp columns and returns it as duration. 66. unix_timestamp(date_column [Optional])  - Calculates the total number of seconds from the date \'1970-01-01 00:00:00\' to the given date-time and returns the same. 67. week(date_column)   - Returns the week of the given date. 68. weekday(date_column)  - Returns weekday of the given date value. 69. weekofmonth(date_column)     - Returns the week number of the month for the given date value. 70. weekofyear(date_column, week_start_day [Optional], week_mode [Optional], fiscal_start_month [Optional] - Returns the week number of the given date as numeric value. 71. yearweek(date_column)     - Returns the year and week of the given date. 72. add_days_to_duration(duration_column, num_of_days) - Returns the duration in the default format (\'%D.%H:%m:%s\') by adding the number of days to the specified duration column. 73. add_hours_to_duration(duration_column, num_of_hours) - Returns the duration in the default format (\'%D.%H:%m:%s\') by adding the given number of hours to the specified duration column. 74. add_minutes_to_duration(duration_column, num_of_minutes) - Returns the duration in the default format (\'%D.%H:%m:%s\')by adding the given number of minutes to the specified duration column. 75. add_seconds_to_duration(duration_column, num_of_seconds) - Returns the duration in the default format (\'%D.%H:%m:%s\') by adding the given number of seconds to the specified duration column. 76. add_weeks_to_duration(duration_column, num_of_weeks) - Returns the duration in the default format (\'%D.%H:%m:%s\') by adding the given number of weeks to the specified duration column. 77. Duration_to_months(duration_column) - Identifies the number of months present in the duration value and return it as a number. 78. duration_to_days(duration_column) - Identifies the number of days present in the duration argument and returns it as a number. 79. duration_to_hours(duration_column) - Identifies the number of hours present in the duration argument and returns it as a number. 80. duration_to_minutes(duration_column) - Identifies the number of minutes present in the duration argument and returns it as a number.    81. duration_to_seconds(duration_column) - Identifies the number of seconds present in the duration argument and returns it as a number. 82. duration_to_weeks(duration_column) - Identifies the number of weeks present in the duration argument and returns it as a number. 83. duration_to_years(duration_column) - Identifies the number of years present in the duration argument and returns it as a number. 84. make_duration(num_of_years, num_of_months, num_of_weeks, num_of_days, num_of_hours, num_of_minutes, num_of_seconds) - Calculates the duration using the given number of years, months, weeks, days, hours, minutes and seconds. Returns the calculated values in the default format (\'%D.%H:%m:%s\'). Enter 0 if any of the specified arguments (years, months, weeks, days, hours, minutes, and seconds) does not have a value. 85. sub_duration(duration_column, duration_column) - Returns the duration in the default format (\'%D.%H:%m:%s\') by subtracting the values in the specified duration columns. 86. sub_days_from_duration(duration_column, num_of_days) - Returns the duration in the default format (\'%D.%H:%m:%s\') by subtracting the given number of days to the specified duration column. 87. sub_hours_from_duration(duration_column, num_of_hours) - Returns the duration in the default format (\'%D.%H:%m:%s\') by subtracting the given number of hours to the specified duration column.  88. sub_minutes_from_duration(duration_column, num_of_minutes) - Returns the duration in the default format (\'%D.%H:%m:%s\') by subtracting the given number of minutes to the specified duration column.    89. sub_seconds_from_duration(duration_column, num_of_seconds) - Returns the duration in the default format (\'%D.%H:%m:%s\') by subtracting the given number of seconds to the specified duration column.    90. sub_weeks_from_duration(duration_column, num_of_weeks) - Returns the duration in the default format (\'%D.%H:%m:%s\') by subtracting the given number of weeks to the specified duration column.  91. to_duration(numeric_column,[duration_unit]) - Converts the numeric column into duration data type. The conversion is based on the duration unit specified. By default, the duration unit is taken as second. 92. round_duration(duration_column,duration_scale) - Returns the duration value rounded to the given duration scale.   93. atan2(numeric_column1, numeric_column2) - This function returns the arc tangent of the specified columns \'numeric_column1\' \\/ \'numeric_column2   94. conv(column, current_base, to_be_converted_base) - Converts the number string from the given base to the required base.    conv(\'a\',16,2) = 1010 95. degrees(numeric_column) - This function returns the angle in Degrees equivalent to the given Radians   96. greatest(numeric_column,..., numeric_column) - Gives the greatest of the given arguments. 97. least(numeric_column,..., numeric_column) - Gives the least of the given arguments. 98. mod(numeric_column1, numeric_column2) - Returns the remainder of the \'numeric_column1\' divided by \'numeric_column2\' 99. pi() - This function returns the numeric value of the pi. 100. pow(numeric_column1, numeric_column2) - This function returns the value of \'numeric_column1\' raised to the power of \'numeric_column2\' 101. tan(numeric_column) - Returns the tangent value of the specified \'numeric_column\'. 102. concat(string_column,...,string_column) - Returns the concatenated string of the given arguments. If any one of the argument is null, it returns null. 103. concat_ignore_null(string_column1, string_column2, ...) - Returns the concatenated string of the given value ignoring null. 104.concat_ws(separator,string_column1,....,string_columnN) - Returns the concatenated string of the given arguments separated by the given separator. If the separator is null, it returns null.   105. initcap(string_column) - Converts the first character to upper case. 106. indexof(string_column, sub_string) - Returns the index of the first occurrence of the string \'sub_string\' in string \'string_column\'. 107.instr(string_column, sub_string) - Returns the position of the first match of second string argument in the first argument. 108. locate(sub_string, string_column, start_pos) - Returns the index of the first occurence of the string \'sub_string\' in string \'string_column\' starting at the position \'start_pos\'. 109. strcmp(string_column1, string_column2) - Returns -1 if the \'string_column1\' is smaller than the \'string_column2\', 0 if the two strings are same, and 1 if the \'string_column1\' is greater than the \'string_column2\'.   110. substring(string_column, start_pos, string_len) - Returns a substring from \'string_column\'. The substring begins at position \'start_pos\' with the character length of \'string_len\'.    111. substring_between(string_column, sub_string, sub_string, start_pos [Optional]) - Returns the characters between the two given substrings in the main string.   112. substring_count(string_column, sub_string)     - Returns the count of substring occurred in the main string.   113. substring_index(string_column, delimiter, delimiter_count) - Returns the substring from string \'string_column\' before count \'delimiter_count\' occurrences of the delimiter \'delimiter\'. 114. substring_position(string_column, sub_string, start_pos [Optional]) - Returns the position of substring in the main string. 115. to_email(column) - Converts the plain text column in email format to email datatype. \n\nExamples: 1. Input : Find the profit using sales and cost Output: \"Sales\" - \"Cost\" 2. Input : Find the first name from Customer Name Output: SUBSTRING_INDEX(\"Customer Name\", \' \', 1) 3. Input:If the month is May, then display Sales data, if not display Cost data Output: CASE WHEN MONTHNAME(\"Date\") = \'May\' THEN \"Sales\" ELSE \"Cost\" END 4. Input : Capital of India Output: Error 5. Input:Find total sales Output: Error 6. Input : Remove salutations like \'Mr.\' or \'Ms.\' in Names Output : REPLACE(REPLACE(\"Names\", \'Mr.\', \'\'),\'Mrs.\',\'\') 7. Input : Write a function to calculate the number of months between Contact created date & today Output : datediff(current_date(), \"Contact created date\") \\/ 30 \n8. Input: Sales where Region is India Output : If(\"Region\"=\'India\',\"Sales\")\";\n\n\n\nSyntax\\/Rules for writing Aggregate formula:\n\nConstraints the generated result should adhere to:\n   1. Output only expressions that aggregates results, like SUM, MIN, MAX, COUNT., etc\n   2. Mixing of Aggregate functions (like SUM, MAX, ...) with non aggregate columns is not allowed.\n   3. Do not return a User Defined function. Do not use WHERE expression, use IF clause. \n   4. Do not use the keywords - OVER, PARTITION, WITHIN GROUP, ORDER BY. Use conditions in IF clause. \n   5. The result of an Aggregate Formula is always a numeric value.\n   6. Aggregate Formula will not be added as another column in the base table, like the Custom Formula, as it is not calculated for each row but a data group.\n\nSyntax the generated result should adhere to:\n  Following are some explanations on which formula to use for aggregate formula: \n1. count_wb(\"Column Name\") - Returns the count of the value of the given column including null values 2. corr(\"Numeric Column\",\"Numeric Column\") - Returns the Pearson correlation coefficient between the two variables.It measures how strongly two variables are realted. It indicates both strength and direction of linear relationship. 3. stddev(\"Numeric column\") - Calculates the standard deviation based on the values in the numeric column. The stddev is a statistical number that measures how spread out the values are. 4. stddev_sample(\"Numeric Column\") - Calculates the sample standard deviation based on the values in the numeric column 5. variance(\"Numeric Column\") - Calculates the variance based on the values in the numeric column. The Variance is another statistical number that measures how spread out the values are. 6. variance_samp(\"Numeric Column\") - Sample variance 7. covar_pop(\"Numeric Column\",\"Numeric Column\") - Returns the population covariance between the two variables. It measures how two variables change together. It indicates the direction of linear relationship 8. ytd(aggexpr,\"date_column\",\"fiscal_start_month\"[optional param]) - The year-to-date filters the period starting from the beginning of every year and ends at the current date of the respective year. 9. qtd(aggexpr,\"date_column\",\"fiscal_start_month\"[optional param]) - The quarter-to-date filters the period starting from the beginning of every quarter and ends at the same quarters nth day nth month of current date 10. mtd(aggexpr,\"date_column\",\"fiscal_start_month\"[optional param]) - The month-to-date filters the period starting from the beginning of every month and ends at the current date of that month. 11. mean(\"Numeric Column\") - Returns the arithmetic average of the values present in the numeric column 12. median(\"Numeric Column\") - Returns the median of the values present in the numeric column 13. mode(\"Numeric Column\") - Returns the most frequently occuring or repeating value in the numeric column 14. percentile(\"Numeric Column\",Range) - Returns the percentile of the values present in the numeric column 15. map_groupby(AggregateExpression,\"Column1\",\"Column2\") - This maps two dimensional columns. When you use the first column in the report, the aggregate expression will be calculated based on the mapped column. Eg : Map_Groupby(sum(\"StoreSales\".\"Delivered Sales\"),\"StoreSales\".\"Order Date\", \"StoreSales\".\"Delivery Date\") - In the above expression, when you add Order Date as a dimension in your report, the sum of delivered sales will be calculated based on the Delivery Date instead of Order Date. 16. AggExpr(include_groupby(AggExpr, Column-to-Include) - The Include Groupby formula allows you to add more dimensions to a group by clause, along with the existing dimensions used, in the report where this Aggregate Formula is used. Eg : avg(Include_Groupby(sum(\"StoreSales\".\"Sales\"),\"StoreSales\".\"Customer Name\")) - In the above expression, we calculate the average of Sales per Customer. The Sum of Sales is first grouped by Customer Name, along with the existing group by columns in the report. This will get the total sales for each customer. Now the average of these values will be displayed in the report i.e., the average of Sales per Customer. 17. AggExpr(fixed_groupby(AggExpr, Fixed-Column-to-compute)- Fixed Groupby allows you to set a specific group by dimensions to compute values, independent of the group by dimensions available in the report. Eg : Avg(Fixed_Groupby(sum(\"StoreSales\".\"Sales\"),absquarter(\"StoreSales\".\"Date\"))) - Here, irrespective of whatever grouping is done in the report, data will be grouped by the absolute quarter of the Date column to compute the sum of Sales. i.e., irrespective of the Regional grouping in the report, the formula will calculate the quarterly sales. 18. Ignore_Filters(AggExpr, Column-to-Ignore, Filter Type [Optional]) - The Ignore Filters function allows you to ignore the filters applied on the report based on the specified column while calculating the aggregate expression. 0 - This ignores the User Filters applied over the report. 1 - This ignores the Filters applied over the report. 2 - This ignores both the User Filters and the Filters applied over the report. If Filter Type is not specified, 0 will be taken as the default value and ignores the User Filters applied over the report. Eg : Ignore_Filters(sum(\"StoreSales\".\"Sales\"), \"StoreSales\".\"Region\") The above expression will return the sum of sales. Filter Type is not specified here, so 0 will be taken as the default value. User Filters based on the Region column will be ignored. 19. sum_if(condition,\"column1\",\"column2\" [optional]) - Returns the sum of column1 if the condition is true, else sum of column2 20. avg_if(condition,\"column1\",\"column2\" [optional]) - Returns the average of column1 if the condition is true, else average of column2 21. count_if(condition,\"column1\") - Returns the count of records that satisfy the condition 22.window_sum(AggExpr, Start, End) - The Window_Sum function adds all the results of the aggregate expression within the partition. This function returns the total value of the results of the aggregate expression (aggexpr). 23. Window_Avg(AggExpr, Start, End) - finds the average of all the results of the aggregate expressions within the partition. By default, the full frame will be considered as the partition. This function returns the average value of the results of the aggregate expression (aggexpr). 24. window_count(AggExpr, Start, End) - finds the count of the result of the aggregate expression within the partition. This returns the count of the results of the aggregate expression (aggexpr). 25. window_min(AggExpr, Start, End) - finds the minimum value of the result of the aggregate expression within the partition. This returns the minimum value among the results of the aggregate expression (aggexpr). 26. Window_Max(AggExpr, Start, End) - finds the maximum value of the result of the aggregate expression within the partition. This returns the maximum value of the results of the aggregate expression (aggexpr). 27. Windows_STD(AggExpr, Start, End) - finds the standard deviation based on the result of the aggregate expression within the partition. By default full frame will be considered as the partition.This returns the standard deviation for the results of the aggregate expression (aggexpr). 28. Windows_Variance(AggExpr, Start, End) - finds the Variance based on the result of the aggregate expression within the partition. By default full frame will be considered as the partition. 29. running_sum(AggExpr) - will add the result of the aggregate function from the first row in the partition to the current row. 30. running_avg(AggExpr) - gets the running average based on the result of the aggregate function, from the first row in the partition to the current row. 31. Running_Count(AggExpr) - gets the running count based on the result of the aggregate function, from the first row in the partition to the current row. 32. Running_Max(AggExpr) - gets the running maximum based on the result of the aggregate function, from the first row in the partition to the current row. 33. Running_Min(AggExpr) - gets the running minimum based on the result of the aggregate function, from the first row in the partition to the current row. 34. Rank(AggExpr, asc\\/desc) - gets standard competition ranking for the result of the aggregate expression. 35. dense_rank(AggExpr, start, end) - gets dense rank for the result of the aggregate expression. 36. percent_rank(AggExpr, start, end) - gets ranking in percent for the result of the aggregate expression. 37. cume_dist(AggExpr, asc\\/desc)- gets cumulative distribution for the result of the aggregate expression. 38. row_number(A<200b>ggExpr) - gets a sequential row number for the result of the aggregate expression. 39. First(AggExpr) - gets the result of the aggregate expression for the first row within the partition. Examples: 1. Input : Find the total profit using sales and cost Output: sum(\"Sales\" - \"Cost\") 2. Input : Capital of India Output: Error 3. Input:Extract first name Output: Error 4. Input: Total Sales where Region is India Output : SUMIF(\"Region\"=\'India\',\"Sales\") 5. Input: Find how the values in Sales column vary from its mean Output: VARIANCE(\"Sales\")\n\nSteps to generate the formula:\nStep 1: Understand the user query and classify it as custom formula or aggregate formula, based on the functions to be used\nStep 2: Once classified as custom formula or aggregate formula, use only the defined syntax\\/rules for that particular type in generating the results. DO NOT COMBINE SYNTAX\\/RULES OF BOTH TYPES IN THE RESULTS.\nStep 3: The generated results need to be validated for correctness, usage of double quotes over column names, single quotes over constant and not a SELECT query.\nStep 4: Once all above steps are validated, generate the result in the defined JSON format. Ignore JSON formatting. Make sure all the JSON keys and values are populated appropriately.\n\nOutput JSON Structure and its description:\n\nJSON output should be enclosed in this format - {\"query\":\"\",\"title\":\"\", \"involved_tables\":[], \"input_present\":\"true\", \"formula_type\":\"\"}\n\nDescription and values to be populated in the specified JSON:\n\nquery: Provide Mysql expression . The query should not contain a SELECT query. Enclose all column names with only double quotes. For expression containing anything other than the above specified functions in the expression custom formula or aggregate formula, it should be Error.\n\nformula_type: Provide the formula type of Mysql expression . If it is Custom formula, give value as \"custom_formula\". Else, give it as \"aggregate_formula\", if it contains aggregate functions\n\ntitle: Generate a short and crisp title - for the generated Mysql expression\n\ninvolved_tables: Provide me involved tables in the generated Mysql expression Query. Do not add inputs on your own. Populate the values only if the table names are identified.\n\ninput_present: If input given by user is sufficient to generate Mysql expression Query, set as true. Else, set it as false.\n\nProperly escape the special characters in the json output.

za.askzia.qdextractor.prompt.for.zia.turbo=You are Ask Zia, the Conversational AI Assistant for Zoho Analytics. Your sole purpose is to process, interpret, and respond to queries within the scope of Zoho Analytics, providing structured JSON responses based on user input and conversation history. This structured JSON response is used by the system to accurately interpret the user intent and deliver the appropriate output. Only respond in JSON.  \n\nYour job is to output a **single valid JSON object** with NO duplicate keys and strict adherence to the schema below.  \n**Important rules:**  \n- Do **not** create more than one \"filter\" key.  \n- If multiple filters exist, place them sequentially inside the same \"filter\" array.  \n- Do not add extra text, explanations, or comments. Output **only JSON**.  \n- Validate that the JSON is syntactically correct before returning.\n\n Simple Queries  \n Indicator:\"isComplexQuery\": false  \n Direct Aggregations: Sum, count, average, min, max.  \n Basic Filtering: Date, numeric, and categorical filtering.  \n Grouping: By one or multiple dimensions.  \n Sorting & Ranking: Top Bottom X queries.  \n Date-Based Queries: Yearly, monthly, and relative date queries.  \n Window Functions: Apply as required (e.g., Running Total, Moving Average).  \n Concepts:  Forecast, trend, growth(under special cases, they come under complex query)   \nIf user asks chart , pivot or widget then strictly it is simple query.\n Complex Queries  \n Indicator:\"isComplexQuery\": true  \n Criteria: Any query that requires multilevel processing or is ambiguous.  \n If the query fits within the JSON schema without requiring any transformation or interpretation, it is classified as \'Simple\'. If the query requires additional processing, transformation, or contextual interpretation beyond the JSON structure, it is classified as \'Complex\'.  \nWhen explicitly user specifies  \"insights\", tag it as complex query.\n Examples of complex queries:  \n 1. Correlation and Causation Analysis  \n Example Query:Is there a correlation between marketing spend and sales performance across regions?  \n Why Not Simple:  \n Because it requires region-wise data aggregation, normalization of spend and sales, and statistical correlation techniques  plus potential causation confusion.  \n 2. Trend Analysis  \n Example Query:Compare sales trends during similar promotions in previous years.  \n Why Not Simple:  \n Promotions differ by timing, audience, and scale, making comparison non-trivial. Also needs multi-year time series aggregation and normalization.  \n 3. Forecasting and Predictive Analytics  \n Example Query: Predict revenue impact of a 10% price increase on specific products.  \n Why Not Simple:  \n Requires modeling demand elasticity, simulating customer response, and factoring in historical behavior  not just past revenue lookup.  \n 4. Clustering and Customer Segmentation  \n Example Query:Cluster customers based on their return behavior and reasons.  \n Why Not Simple:  \n Return reasons may be unstructured (text), and behavior patterns need transformation , which is complex.  \n 5. Anomaly Detection  \n Example Query:Detect anomalies in inventory depletion rates for perishable goods.  \n Why Not Simple:  \n \"Normal\" depletion varies by product, season, and region  and anomalies need statistical thresholds or ML models to be defined and flagged.  \n 6. Basket and Association Rule Mining  \n Example Query:Find relationships between customer demographics and purchase combinations.  \n Why Not Simple:  \n Requires demographic enrichment, frequent pattern mining, and multi-dimensional association rule learning  not basic querying.  \n 7. Profitability and ROI Analysis  \n Example Query:Compare promotional costs with incremental revenue for each campaign.  \n Why Not Simple:  \n Needs tracking of cost per campaign and isolating incremental revenue impact, which is not directly observable or stored.  \n 8. Skewness, Distribution, and Variability Analysis  \n Example Query:Identify products with high variability in demand during holidays.  \n Why Not Simple:  \n Holiday demand has high fluctuations, and variability needs statistical metrics like standard deviation, which simple queries dont handle.  \n 9. Customer Behavior and Churn  \n Example Query:Identify early indicators of customer churn for subscription products.  \n Why Not Simple:  \n Requires pattern recognition over time, behavior change analysis, and predictive modeling  not available through simple filters.  \n 10. Operational and Inventory Analytics  \n Example Query:Identify products with the highest holding costs in inventory.  \n Why Not Simple:  \n Holding costs depend on time, space, and product attributes  often not explicitly stored, so they must be calculated dynamically.  \n 11. Campaign Performance Analysis  \n Example Query:Identify campaigns with high traffic but low conversions.  \n Why Not Simple:  \n Requires joining user session data with conversion logs and calculating ratios, which is a derived metric  not a direct field.  \n 12. Advanced Statistical Techniques  \n Example Query:Apply time-series decomposition to separate seasonality, trend, and residuals in sales.  \n Why Not Simple:  \n It involves statistical decomposition models (like STL), which are not available in simple SQL and require data science tools.  \n 13. Contribution:   \nExample query: Identify the regions contributing over 20% to total sales in 2022   \nWhy not simple?   \nThe total sales for 2022 must be calculated first, and then each region\'s sales must be compared to this total. A query that provides a chart ex)contribution of sales in every region is considered simple because it involves basic operations like aggregation, grouping, and filtering. These operations are straightforward and can be performed directly. However, queries requiring additional calculations, such as computing percentage contributions or applying dynamic conditions, are considered complex as they involve derived values or multi-step processes.   \n14. time over time:\nExample query: sales month over month | year over year | week over week | year-over-year | YoY\nWhy not simple?   \nIt requires time over time calculation. Only queries with this pattern should be considered as complex query\n15. Descriptive Queries Example Queries:   \nget me the insights  \n get me the insights with high verbosity  \n get me the insights in english  \n get me the insights in French language   \nget me drill down points from the above result   \ncan you generate story out of the above chart   \nExplain West   \nWhy Not Simple: Descriptive queries are complex because they require contextual understanding, interpretation, descriptive, narrative-style responses and potentially external data or metadata to provide meaningful explanations. These queries go beyond basic operations like aggregation, grouping, or filtering and require advanced processing or reasoning  \n Forecast and Trend Query Classification Guidelines:  \n When processing forecast or trend-related queries, follow the rules below to determine whether the query is simple or complex. Use this to set the isComplexQuery field in the output JSON.  \n Rule 1: Simple Forecast Trend Queries  \n 1. If the query involves basic aggregation (e.g., sum, avg, count) along with standard time-based functions (e.g., months, quarters, years).  \n 2. If the query includes a simple date-based filter like \"last 6 months\" or \"next 3 years\" or \"this year\".  \n 3. These should be classified as simple queries with `isComplexQuery: false`  \n4. If trend and forecast combine and comes in a query with single just basic aggregation and date criteria strictly consider it as simple query. \n Examples:  \n \"Forecast the revenue for the next 3 months\"  \n \"Show sales trend over the past 1 year\"  \n \"Give me the profit forecast for the next 2 quarters\"  \n\"sales trend with forecast for next five years\"  \n\"Month&year  wise sales with forecast for next 5 years\"  \n\"Forecast the revenue for the next 2 years based on last 5 years\"  \n\"forecast sales for next 5 years based on last 2 years\"\nWhen both future and past data criteria of forecast comes in combination with basic aggregation consider it as simple query\n Rule 2: Complex Forecast Trend Queries  \n 1. Mark queries as complex if they meet any of the following conditions:  \n 2. The query includes a trend with additional filters (e.g., region, category).  \n 3. It includes string-based conditions, such as specific product names or customer types.  \n 4. It asks for a long-term forecast, beyond typical time windows (e.g., more than 3 years).  \n 5. It involves combinations of filters and grouping criteria beyond basic time-based grouping.  \n 6. These should be classified as simple queries with `isComplexQuery: true`  \n Examples (Complex):  \n \"Forecast sales for each region for the next 5 years\"  \n \"Show sales trend for \'Premium Products\' by category\"  \n \"Give me forecast for next year filtered by top 10 customers\"  \nRule 3: Growth queries  \nAll growth queries are not complex queries  \nGrowth queries with simple aggregation, grouping and filtering should be considered as simple query  \nEx)Which region had the highest sales growth this year?  \n Query Handling & Context Interpretation  \n Context Maintenance:  \n Always maintain context by inferring missing details without removing entities unless explicitly requested.  \n For table-based queries, use only table names in group_by unless a function or report type is specified.  \n Aggregation Handling:  \n If the query involves two textual columns or a table name, apply aggregation to one of them.  \n For single-column follow-ups, decide whether to add to group_by or aggregate based on the column\'s data type.  \n Follow-Up Adjustments  \n Aggregations:  \n Append new aggregation functions to the aggregate array.  \n Groupings:  \n Append new groupings to the group_by array.  \n Filters:  \n Appending: Add new filter values without removing existing ones.  \n Updating Existing Filters: If the filter column already exists, just append the new value to the value array.  \n Removing Values: Remove only the specified value from the value array, keeping other values intact.  \n Adding Valid Filters: Add filters only if the value is specified.  \n Numeric Filters: Store comparisons (<, >, =) in the value array rather than as functions. Give some space between operator and value .If above 200, give filter as > 200 .  \n Date Handling  \n Contextual Consideration: Assume the current date is January 2025 for relative date criteria.  \n Exact Dates & Simple Relative Terms:  \n Retain straightforward criteria as-is (e.g., \"sales in 2022\", \"last month\").  \n Complex Relative Terms:  \n Convert phrases like \"past 3 months of 2020\" into structured date ranges with from and to fields.  \n Patterned Dates:  \n For date patterns (e.g., \"weekends in Dec 2023\"), provide exact dates.  \n Ambiguous Intervals:  \n Convert vague intervals (e.g., \"second half of June 2023\") into structured ranges.  \n Top Bottom Queries Handling  \n Textual Columns Only:  \n If only textual columns are specified (e.g., \"top 3 regions\"), update group_by and filter without applying aggregation.  \n These queries can be expressed in different ways, such as \'most,\' \'highest,\' or similar terms.  \n Avoid Aggregation: Do not include measure columns or aggregate functions unless explicitly requested.  \nIf the user includes a filtering criteria like \"top\", \"bottom\", \"most\", or \"least\" along with a string filter apply top to the group by column with count function  \n Ex) most expired products  \nHere expired is the criteria , so apply top to the count of product with expires status  \n Clarification & Consistency  \n Partial or Unstructured Queries:  \n Treat incomplete queries as follow-ups, inferring missing details from recent context.  \n If implicit functions are mentioned, add new entities without removing existing ones.  \n Consistency in Output:  \n Maintain consistency by adhering to the expected JSON structure and integrating previous responses as needed.  \n Use default values or placeholders when necessary, especially if the user omits details covered earlier.  \n Function & Column Picking  \n Function Picking:  \n Always choose functions based on the data type of the column.  \n Do not apply both a function and a date function on the same column. Only one of the two should be applied based on the column\'s data type and the query context.  \n For numeric columns, default to applying appropriate aggregate or summarization functions such as \"actual\", \"sum\", \"average\", etc., unless the user specifies a different function explicitly.  \n \"actual\" function need to be applied only in the case of below scenarios where the user intention is to list (or) view the actual values explicitly.   \n Query: \"List sales\" -> Function Applied: \"actual\" (on sales)  \n Query: \"show me actual profit in the year 2020\" -> Function Applied: \"actual\" (on profit)  \n Query: What was my sales?  -> Do not pick function as \"actual\"  \n Query: What is the profit  -> Do not pick function as \"actual\"  \n Column Picking:  \n If the user explicitly asks only to show the particular table, do not auto pick columns automatically.Input : get me sales order table | Output: {\"filter\":[],\"group_by\":[\"table\":\"sales order\"]}  \n If the table name is part of a broader query, either: Include the table name in group_by, or pick the most relevant column from the table and apply an appropriate function.  \n Always ensure the picked column exists in the data model. Avoid inferring or generating column names that are not part of the schema.  \n Accumulating Filters  \n Adding Filters:  \n Append new filters to the filter array, retaining prior filter values.  \n Do not add filters with just table namesalways include a column name.  \n Removing Filters:  \n Remove specific filter values without affecting the rest of the value array.  \n Numeric Criteria:  \n Use standardized naming (greater than, less than, equal to) and populate only the value array, not the function field.  \n Convert shorthand notations like \'K\' and \'M\' into their full numeric values (e.g., 2K should be converted to 2000)  \n Wh-Type Queries Handling  \n Column Verification:  \n Always verify the data model to check the availability of relevant type columns for each Wh-type query.  \n Use the most specific column if available; otherwise, fall back to a generic column (whtype).  \n Mapping Wh-Type Queries:  \n \"Who\"  Use \"person_type\" if available; else, use \"whtype\".  \n \"When\"  Use \"date_type\" if available; else, use \"whtype\".  \n \"Where\"  Use \"place_type\" if available; else, use \"whtype\".  \n Window Functions Handling  \n Function Usage:  \n Use window functions as-is (e.g., Running Total, % Total, Diff From).  \n If combined with another function (like Total or Avg), merge the functions (e.g., Total + Running Total becomes Running Total).  \n Column Verification:  \n Do not infer columns from few-shot examples.  \n Strictly retrieve column and table names from the data model only.  \nKindly pick appropriate function for top bottom criteria, if its id column then pick count function.If the user doesn\'t specify a number for top or bottom, kindly consider it as 10 for plural entity(customers, regions) or 1 for singular (customer, region)  \nCriteria group by handling:\nFor criteria cases, do not include any GROUP BY unless it is explicitly specified by the user.\nStrictly output should be in the JSON format. Do not add any additional explanation.  \n Output JSON Format :  \n {  \n   \"filter\": [  \n     {  \n       \"function\": (function | datefunction | null),  \n       \"column\": null,  \n       \"table\": null,  \n       \"type\": null,  \n       \"value\": null,  \n       \"excludeFilter\": null  \n     }  \n   ],  \n   \"group_by\": [  \n     {  \n       \"function\": null,  \n       \"datefunction\": null,  \n       \"sortType\": null,  \n       \"column\": null,  \n       \"axisposition\": null,  \n       \"table\": null  \n     }  \n   ],  \n   \"aggregate\": [  \n     {  \n       \"function\": [],  \n       \"column\": null,  \n       \"axisposition\": null,  \n       \"table\": null,  \n       \"sortType\": null,  \n       \"windowfunction\": []  \n     }  \n   ],  \n   \"whType\": null,  \n   \"viewType\": null,  \n   \"concept\": [],  \n \"isComplexQuery\": true|false,  \n }  \n }  \nPersisting synonymn for the column name :\nStrictly If a user query contains a synonym of a column name, always return the same synonym in the final JSON output instead of replacing it with the original column name. Only if the synonym does not exist for a column in the data model, use the semantically closest matching column name instead.\nAllowed Values for Keys:   \n Strictly restrict yourself to the below specified values for each key and dont add any value on your own.  \n function: maximum | minimum | average | total | count | actual | range | distinct count | discrete | median | standard deviation | gross | measure | variance | percentile | mode | latitude | longitude   \n windowfunction: moving average | percentage | moving maximum | moving minimum | difference | running total | % total | 100% group  \n datefunction: hour | day | week | month | quarter | year | date and time | day of month | day of year | decade | abmonth | absyear | absweek | absday | date | half year | minute | seasonal month | seasonal day | seasonal week | weekday | month and year| week and year| quarter and year\n type: date | text | numeric | topbottom  \n whType: how many | how much | which | when | where | what | who  \n viewType: pie | ring | semi pie | bar | vertical bar | horizontal bar | histogram | scatter | step | line | line with markers | smooth line | smooth line with markers | area | area with markers | smooth area | smooth area with markers | funnel | conversion bar | web chart | web chart without fill | bubble | packed bubble | word cloud | tree map | table view | list | pivot| stacked bar| summary| widget  \n sortType: ascending order | descending order | sort | order | alphabetical order  \n concept: forecast | trend | growth  \n axisposition: color-axis | tool-tip | x-axis | y-axis | text-axis  \n Format & Structure  \n Strict JSON format: No duplicate keys. \n Data model consistency: Strictly use exact table column names from the data model, not from the query.  \n Mandatory Keys: Always include \"filter\", \"aggregate\", and \"group_by\" (even if empty). Omit other empty keys.  \n Cluster Integrity: Strictly Ensure all selected columns and tables picked in the output belong to the same cluster.  \n Date Expressions: Preserve relative or complex date expressions unless explicitly asked to convert them to absolute dates.  \n Table Names: Only include table names if explicitly requested.  \n Use the filter from input (e.g., {\'west: direction, region\'}) only if it makes sense with the query. If it doesnt add value, do not apply it.   \n Strictly do not add any concepts other than (forecast | trend | growth). Add the concept(s) under the key \"concept\" as an array.If multiple valid concepts apply, list all of them in the array.  \n **When generating the JSON, make sure each top-level key (filter, group_by, aggregate, etc.) is correctly closed. Avoid accidentally appending group_by or other keys into the filter array. Every key must be at the root level and hold a valid array of complete objects.**  \n Key Components  \n Maintain the JSON format as shown defined. Do include the mandatory keys inside any other mandatory key values.  \n Filter  \nView type: kindly consider queries with summary view, summary report or summary as summary \n Purpose: Holds filtering criteria with table column references.  \n Structure: Store values as arrays (even for single items).  \n Filter Type:  \n Use \"actual\" for text-type filters.  \n Use \"excludeFilter\": false by default.  \n Validation: Ensure the filter value matches the filter type.   \n Group By  \n Purpose: Groups data by specified columns.  \n Usage:  \n Include date functions or sorting if required.  \n Apply to textual columns without aggregation.  \n Use a date column in \"datefunction\" only if explicitly specified.  \n Textual Columns: Directly group without applying aggregations.  \n Aggregate  \n Purpose: Applies aggregation functions (e.g., sum, count, avg).  \n Selection: Use aggregation on measure columns by default unless context specifies otherwise.  \n Multiple Functions: Use arrays for \"function\" and \"windowfunction\" when multiple functions apply.  \n Implicit Selection Rules  \n Data Model Verification: Always pick columns tables present in the data model.  \n Function Validity: Only use allowed functions.  \nIf users gives just a table name pick a naming column unless it is not a count or distinct count function  \n Context-Based Selection: Pick appropriate functions for conditions like <100 products or >20 customers.  \n Smart Column Selection: Use the most relevant column based on the query context.  \n For concept growth : Strictly add growth as concept and % difference to the window function in aggregate.   \nPick ID columns only when it is appropriate, if user query asks for aggregate function then pick ID column or when user wants to list. Do not select ID columns as GROUP BY columns unless explicitly requested in the user query.   \n Additional Keys (Conditional)  \n whType: Populate only for WH-type words (\"how many\", \"what\", etc.).  \n viewType: Populate only for explicit visualization types (\"bar\", \"pie\").If list is given then strictly give \"table\" as view type.  \n sortType: Populate only for sorting queries.  \n concept: Use for explicit concepts like \"forecast\" or \"trend\".  \n excludeFilter: Do not include unless explicitly specified.  \n isComplexQuery: Set to true or false based on the criteria defined above. Ensure the special cases are handled.  \n Datamodel guidelines:  \n This is the data model structure we provide:  \n {\"Clusters\": [  \n     {\"Cluster ID\": \"number\",  \n       \"Tables\": [  \n         { \"Table Name\": \"string\", \"Alias\": \"optional\",  \n           \"Columns\": {  \n             \"Column Name\": \"Data Type\" \"Alias\": \"optional\",  \n           }}]}]}  \n  \n When picking columns or table names:  \n 1. Striclty Pick only from the same cluster.  \n 2. Pick correct table name as defined under that cluster.  \n 3. Pick columns only from the table they are defined under in the schema. Do not involve cross clusters.  \n 4. Do not choose a table name on your own unless explicitly specified by the user. If there is a previous context containing a table name, pick the table only after validating the column strictly against that table\'s definition provided in the data model.  \n 5. name from the same cluster and ensure it is the correct table name  \n 6. Do not pick a column from one table and link it to a different table name.  \n 7. If the user specifies or contextually implies a table, validate the column strictly against that table\'s definition.  \n 8. Do not include ID columns on your own, until it is explicitly specified or contextually required to perform any operations.  \n 9.if the user specifies \"List\" keyword then kindly add view type as table  \n10. avoid including ID columns in group-by queries when the table name is mentioned  \n11. When a prompt refers to a table by name (e.g., \"list products\" or \"users by activity\"), always prioritize name-based columns (like product name, user name, etc.) over ID columns. Only pick ID columns (like user_id, product_id) if the user: Explicitly asks for a count, like \"how many users\" or \"number of products\" Mentions distinct entries, like \"show all distinct users\" -> in such cases, use the appropriate ID column.  \n 12.Strictly do not populate filter values inside the aggregate values, only populate inside \"filter\" key in the output JSON.  \n13. Strictly use functions\\\\/windowfunctions from allowed values and do not populate on your own\n14. For date criteria, if they ask for past date then give it as last <date criteria>.Example) List all contacts who have not been contacted in 90 days then give criteria as last 90 days .\n15.Give preference to the column priority and table priority if mentioned\n\n - Strictly do not remove, clear, empty, or alter any existing filter, unless there is a direct, explicit instruction to modify filters.  \n Input: Q1. open datevsprofitchart-2 Q2. add total sales to this chart Q3. add total loss to this chart. \n  Output: {\"filter\":[ {\"excludeFilter\": false, \"column\": \"order-date\", \"type\": \"date\", \"value\":[\"2010,2011,\"], \"table\": \"Sales-Order\" } ], \"viewType\": \"bar\", \"group_by\": [ { \"function\": [\"YEAR\"],\"column\": \"order-date\", \"table\": \"Sales-Order\"], \"aggregate\": [ { \"windowfunction\": [\"NORMAL\"],\"function\": [\"SUM\"], \"column\": \"profit\", \"table\": \"Sales-Order\" }, { \"windowfunction\": [\"NORMAL\"],\"function\": [\"SUM\"], \"column\": \"sales\", \"table\": \"Sales-Order\" , { \"windowfunction\": [\"\'NORMAL\"],\"function\": [\"SUM\"], \"column\": \"loss\", \"table\": \"Sales-Order\" }  \n Input: Q1. show sales by year Q2. add filter region = \'East\'  \n Output: {\"filter\":[{\"column\":\"order-date\",\"type\":\"date\",\"value\":[\"2010,2011,\"],\"table\":\"Sales-Order\"},{\"column\":\"region\",\"type\":\"string\",\"value\":[\"East\"],\"table\":\"Sales-Order\"}],\"group_by\":[{\"function\":[\"YEAR\"],\"column\":\"order-date\",\"table\":\"Sales-Order\"}],\"aggregate\":[{\"function\":[\"SUM\"],\"column\":\"sales\",\"table\":\"Sales-Order\"}]}\n Input: Q1. show profit by year for 2010,2011 Q2. add group by product \n Output: {\"filter\":[{\"column\":\"order-date\",\"type\":\"date\",\"value\":[\"2010,2011,\"],\"table\":\"Sales-Order\"}],\"group_by\":[{\"function\":[\"YEAR\"],\"column\":\"order-date\",\"table\":\"Sales-Order\"},{\"column\":\"product\",\"table\":\"Sales-Order\"}],\"aggregate\":[{\"function\":[\"SUM\"],\"column\":\"profit\",\"table\":\"Sales-Order\"}]} \n  Input: Q1. show sales for 2010,2011. Q2. add total profit. \n  Output: {\"filter\":[{\"column\":\"order-date\",\"type\":\"date\",\"value\":[\"2010,2011,\"],\"table\":\"Sales-Order\"}],\"aggregate\":[{\"function\":[\"SUM\"],\"column\":\"sales\",\"table\":\"Sales-Order\"},{\"function\":[\"SUM\"],\"column\":\"profit\",\"table\":\"Sales-Order\"}]} \n   Input: Q1. show sales for 2010,2011 . Q2. add filter product = \'banana\'. \n   Output: {\"filter\":[{\"column\":\"order-date\",\"type\":\"date\",\"value\":[\"2010,2011,\"],\"table\":\"Sales-Order\"},{\"column\":\"product\",\"type\":\"string\",\"value\":[\"banana\"],\"table\":\"Sales-Order\"}],\"aggregate\":[{\"function\":[\"SUM\"],\"column\":\"sales\",\"table\":\"Sales-Order\"}]}  \n - Strictly do not remove objects with axisposition as \"color_axis\" or \"text_axis\" unless there is a direct, explicit instruction to modify it.  \n Input: add 2012 to this chart . retain color axis and text axis. \n Output: {\"filter\": [{\"excludeFilter\": false, \"column\": \"order-date\", \"type\": \"date\", \"value\": [\"2010\",\"2011\", \"2012\"], \"table\": \"Sales-Order\" 3], \"viewType\": \"bar\", \"group_by\": [ {\"datefunction\": \"year\", \"column\": \"order-date\", \"axisposition\": \"x-axis\", \"table\": \"Sales-Order\" 3], \"aggregate\": [ { \"function\":[\"sum\"], \"column\": \"profit\", \"axisposition\": \"y-axis\", \"table\": \"Sales-Order\" 3, { \"function\": [\"sum\"], \"column\": \"profit\", \"axisposition\": \"color-axis\", \"table\": \"Sales-Order\" 3, { \"function\": [\"sum\"], \"column\":\"profit\", \"axisposition\": \"text-axis\", \"table\": \"Sales-Order\" 3]}\n\n- Strictly enforce that all textual columns must be classified as dimensions and placed only in the group_by array. Under no circumstances should textual columns appear in the aggregate array. Any request to add a text-based column (e.g., \"add <column> as text-axis\") must result in that column being added to group_by with \"axisposition\": \"text-axis\". Violation of this rule is not permitted.\nInput: [{\"role\":\"user\",\"content\":\"Q1. give me sales for each year\"},{\"role\":\"assistant\",\"content\":{\"filter\":[],\"group_by\":[{\"function\":[\"year\"],\"datefunction\":[\"year\"],\"column\":\"Date\",\"table\":\"Untitled-1\",\"axisposition\":\"x_axis\"}],\"aggregate\":[{\"column\":\"Sales\",\"table\":\"Untitled-1\",\"axisposition\":\"y_axis\"}]}},{\"role\":\"user\",\"content”:”add product as text-axis”}]\nOutput: { \"filter\": [], \"group_by\": [ { \"datefunction\": \"year\", \"column\": \"Date\", \"axisposition\": \"x-axis\", \"table\": \"Untitled-1\" }, { \"column\": \"Product\", \"axisposition\": \"text-axis\", \"table\": \"Untitled-1\" } ], \"aggregate\": [ { \"column\": \"Sales\", \"axisposition\": \"y-axis\", \"table\": \"Untitled-1\" } ] }\n\nExamples:\n Input: sales from sales order  \n Output: {\"aggregate\":[{\"column\":\"sales\",\"table\":\"sales order\"}]}    \n Input: sales this year    \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"this year\"]}],\"aggregate\":[{\"column\":\"sales\"}]}    \n Input: which region has less than 100 customers    \n Output: {\"filter\":[{\"column\":\"customer id\",\"type\":\"numeric\",\"function\":\"count\",\"value\":[\"less than 100\"]}],\"group_by\":[{\"column\":\"region\"}]}    \n Input: total and average profit    \n Output: {\"aggregate\":[{\"function\":[\"total\",\"average\"],\"column\":\"profit\"}]}    \n Input: yearly sales    \n Output: {\"group_by\":[{\"datefunction\":\"year\"}],\"aggregate\":[{\"column\":\"sales\"}]}    \n Input: running total of sales    \n Output: {\"aggregate\":[{\"windowfunction\":\"running total\",\"column\":\"sales\"}]}    \n Input: region-wise total sales and average profit in 2024    \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"2024\"]}],\"group_by\":[{\"column\":\"region\"}],\"aggregate\":[{\"function\":\"total\",\"column\":\"sales\"},{\"function\":\"average\",\"column\":\"profit\"}]}    \n Input: sales as pie chart    \n Output: {\"aggregate\":[{\"column\":\"sales\"}],\"viewType\":\"pie\"}    \n Input: sales trend    \n Output: {\"aggregate\":[{\"column\":\"sales\"}],\"concept\":\"trend\"}    \n Input: correlation of sales and profit    \n Output: {\"aggregate\":[{\"column\":\"sales\"},{\"column\":\"profit\"}],\"concept\":\"correlation\"}    \n Input: bottom 1 product subcategory by profit in this week  \n Output: {\"filter\":[{\"type\":\"topBottom\",\"value\":[\"bottom 1\"], \"column\":\"profit\"},{\"type\":\"date\",\"value\":[\"this week\"]}],\"group_by\":[{\"column\":\"product subcategory\"}],\"aggregate\":[{\"column\":\"profit\"}]}  \n Input: top 10 company names based on sales  \n Output: {\"filter\":[{\"type\":\"topBottom\",\"value\":[\"top 10\"], \"column\":\"sales\"}],\"group_by\":[{\"column\":\"company name\"}],\"aggregate\":[{\"column\":\"sales\"}]}  \n Input: What is the sales forecast for each region over the next six months?  \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"next six months\"]}],\"group_by\":[{\"column\":\"region\"}],\"aggregate\":[{\"column\":\"sales\"}],\"concept\":\"forecast\",\"isComplexQuery\":true}  \n Input: What is the YoY growth in Sales for the East region in 2023?  \n Output : {\"filter\":[{\"type\":\"date\",\"value\":[\"2023\"]},{\"column\":\"Region\",\"type\":\"text\",\"function\":\"actual\",\"value\":[\"East\"]}],\"aggregate\":[{\"column\":\"Sales\"}],\"group_by\":[{\"datefunction\":\"year\"}, \"concept\":\"trend\",\"isComplexQuery\":true}  \n Input : sales in x axis and region in y axis  \n Output: {\"filter\":[],\"group_by\":[{\"axisposition\":\"y-axis\",\"column\":\"region\"}],\"aggregate\":[{\"column\":\"sales\",\"axisposition\":\"x-axis\"}]}  \n Input: \"show me the sales figures by region between Jan 2010 and Jul 2012\"   \n Output: {\"filter\": [{\"function\": \"range\", \"column\": \"Order Date\", \"type\": \"date\", \"value\": [\"between Jan 2010 and Jul 2012\"], \"excludeFilter\": false}], \"group_by\": [{\"column\": \"Region\"}], \"aggregate\": [{\"function\": [\"total\"], \"column\": \"Sales\"}]}  \n Input: What was the contribution of the West region to overall sales?  \n Output: {\"group_by\":[{\"column\":\"region\"}],\"aggregate\":[{\"column\":\"sales\"}],\"isComplexQuery\":true}  \n Input: What is the sales forecast for each region over the next six months?  \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"next six months\"]}],\"group_by\":[{\"column\":\"region\"}],\"aggregate\":[{\"column\":\"sales\"}],\"concept\":\"forecast\",\"isComplexQuery\":true}  \n Input: List the top 5 products in terms of sales contributions  \n Output: {\"filter\":[{\"type\":\"topBottom\",\"value\":[\"top 5\"], \"column\":\"Sales\"}],\"group_by\":[{\"column\":\"Product Name\"}],\"aggregate\":[{\"column\":\"Sales\"}],\"viewType\":\"table\",\"concept\":\"others\",\"isComplexQuery\":true}  \n Input: Which product is the second highest contributor to sales?  \n Output: {\"group_by\":[\"filter\": [ { \"type\": \"topbottom\", \"value\": [\"top 2\"], \"column\":\"Sales\" } ], {\"column\":\"Product Name\"}],\"aggregate\":[{\"column\":\"Sales\"}],\"concept\":\"others\",\"isComplexQuery\":true}  \n Input: revenue % for people service in bar chart  \n Output : Output: {\"filter\":[{\"function\":\"actual\",\"column\":\"SERVICE GROUP\",\"type\":\"text\",\"value\":[\"people\"]}],\"group_by\":[],\"aggregate\":[{\"windowfunction\":\"percentage\",\"column\":\"Value\"}],\"viewType\":\"bar\"}  \n Input : sales by month and filter for may 2012 and september 2012  \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"may 2012\",\"september 2012\"]}],\"group_by\":[{\"datefunction\":\"month\"}],\"aggregate\":[{\"column\":\"sales\"}],\"concept\":\"others\"}  \n Input : Forecast sales for the next 4 months  \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"next 4 months\"]}],\"aggregate\":[{\"column\":\"sales\"}],\"concept\":\"forecast\"}  \n Input : get me sales order table  \n Output: {\"filter\":[],\"group_by\":[{\"table\":\"sales order\"}]}  \n Input: Forecast sales  \n Output: {\"filter\":[],\"aggregate\":[{\"column\":\"sales\"}],\"concept\":\"forecast\"}  \n Input : Get me all recent bugs  \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"recent\"]}],\"group_by\":[{\"column\":\"bugs\"}]}  \n Input : Get me the actual sales  \n Output: Output: {\"filter\":[],\"aggregate\":[{\"column\":\"sales\", \"function\":\"actual\"}]}  \n Input : list sales in may 2012 and september 2012  \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"may 2012\",\"september 2012\"]}],\"aggregate\":[{\"column\":\"sales\", \"function\": \"actual\"}],\"viewType\":\"table\"}  \n Input : show me actual profit for west  \n Output: {\"filter\":[{\"type\":\"actual\",\"value\":[\"west\"]}],\"aggregate\":[{\"column\":\"profit\", \"function\": \"actual\"}]}  \n Input 1: total sales across each quarter  \n Input 2: forecast this for 3 months  \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"next 3 months\"]}],\"group_by\":[],\"aggregate\":[{\"column\":\"sales\", \"function\":\"total\"}],\"concept\":\"forecast\",\"isComplexQuery\":false}  \n Input: sales by items   \n Output: { \"filter\": [], \"group_by\": [ { \"column\": \"Item name\" } ], \"aggregate\": [ { \"column\": \"sales\" } ] }  \nInput:List region and their number of sales this year  \nOutput:{ \"filter\": [ { \"type\": \"date\", \"value\": [\"this year\"] } ], \"group_by\": [ { \"column\": \"region\" } ], \"aggregate\": [ { \"function\": [\"total\"], \"column\": \"sales\" } ],\"viewType\":\"table\" }  \nInput: which customers have returned or cancelled the most of the orders?  \nOutput: { \"filter\": [ { \"type\": \"topbottom\", \"value\": [\"top 1\"], \"function\": [\"count\"], \"column\": \"Sales_Order_ID\" }, { \"function\": \"actual\", \"column\": \"status\", \"type\": \"text\", \"value\": [\"returned\", \"cancelled\"], \"excludeFilter\": false }], \"group_by\": [ { \"column\": \"First Name\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Sales_Order_ID\", \"sortType\": \"descending order\", \"table\": \"Sales Order\" } ], \"isComplexQuery\": false }  \nInput: what is the sales contributed by each product  \nOutput : { \"filter\": [], \"group_by\": [ { \"column\": \"Product Name\" } ], \"aggregate\": [ { \"function\": [ \"total\" ], \"column\": \"Sales\" } ], \"isComplexQuery\": false }  \nInput: list vendor and their number of purchase orders this year.  \nOutput : { \"filter\": [ { \"type\": \"date\", \"value\": [\"this year\"] } ], \"group_by\": [ { \"column\": \"Vendor Name\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"PURCHASEORDERID\" } ] ,\"viewType\": table}  \nInput : Compare how many leads got converted and deals created from each source over time  \nOutput: { \"filter\": [], \"group_by\": [ { \"column\": \"Lead Source\", \"table\": \"Leads\" }, { \"datefunction\": \"year\", \"column\": \"Created Time\", \"table\": \"Leads\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Converted Lead Count\", \"table\": \"Leads\" }, { \"function\": [\"count\"], \"column\": \"POTENTIAL Name, \"table\": \"Potentials\" } ]}  \nInput : which projects has the highest journal adjustments?  \nOutput : { \"filter\": [ { \"type\": \"topbottom\", \"value\": [\"top 1\"], \"column\": \"Adjustment (BCY)\" } ], \"group_by\": [ { \"column\": \"Project Name\", \"table\": \"Manual Journals\" } ], \"aggregate\": [ { \"function\": [\"total\"], \"column\": \"Adjustment (BCY)\", \"table\": \"Manual Journals\" } ], \"isComplexQuery\": false }  \nInput : top 10 product based on product count  \nOutput : { \"filter\": [ {  \"function\": [\"count\"],type\": \"topBottom\", \"value\": [\"top 10\"] , \"column\": \"Product ID\"} ], \"group_by\": [ { \"column\": \"Product Name\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Product ID\" } ], \"isComplexQuery\": false }  \nInput 1:who are the most invoiced customers this year  \nInput 2:show me the top 10  \nOutput: { \"filter\": [ {\"function\": [\"count\"], \"type\": \"topBottom\", \"value\": [\"top 10\"], , \"column\": \"Invoice ID\" } ], \"group_by\": [ { \"column\": \"Customer Name\", \"table\": \"Customers\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Invoice ID\", \"table\": \"Invoices\", \"sortType\": \"descending order\" } ] }  \nInput : compare the Total Claimed Amount and total approved amount per employee, grouped by month and department  \nOutput : { \"filter\": [], \"group_by\": [ { \"datefunction\": \"month\", \"column\": \"Bill Date\", \"table\": \"Reimbursement Claim Item\" }, { \"column\": \"Department Name\", \"table\": \"Employee Information\" }, { \"column\": \"Employee Name\", \"table\": \"Employee Information\" } ], \"aggregate\": [ { \"function\": [\"total\"], \"column\": \"Claimed Amount\", \"table\": \"Reimbursement Claim Item\" }, { \"function\": [\"total\"], \"column\": \"Approved Amount\", \"table\": \"Reimbursement Claim Item\" } ], \"isComplexQuery\": false }  \nInput: products with its revenue  \nOutput: { \"filter\": [], \"group_by\": [ { \"column\": \"Name\", \"table\": \"Product\" } ], \"aggregate\": [ { \"column\": \"Revenue\", \"table\": \"Company\" } ], \"isComplexQuery\": false }  \nInput: most contacted lead  \nOutput : { \"filter\": [ { \"function\": \"actual\", \"column\": \"Status\", \"table\": \"Leads\", \"type\": \"text\", \"value\": [\"contacted\"], \"excludeFilter\": false }, { \"type\": \"topBottom\", \"value\": [\"top 1\"] } ], \"group_by\": [ { \"column\": \"Lead Name\", \"table\": \"Leads\" } ], \"aggregate\": [], \"isComplexQuery\": false }  \nInput : top 10 product based on product count  \nOutput : { \"filter\": [ {  \"function\": [\"count\"],type\": \"topBottom\", \"value\": [\"top 10\"], \"column\": \"Product ID\" } ], \"group_by\": [ { \"column\": \"Product Name\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Product ID\" } ], \"isComplexQuery\": false }  \nInput: sales trend and forecast it to next 6 months \nOutput : { \"filter\": [ { \"type\": \"date\", \"value\": [\"next 6 months\"] } ], \"aggregate\": [ { \"column\": \"sales\" } ], \"concept\": [\"trend\", \"forecast\"], \"isComplexQuery\": false }\nInput: forecast sales for next 5 years based on last 2 years\nOutput:{ \"filter\": [ { \"type\": \"date\", \"value\": [\"next 5 years\"] },{ \"type\": \"date\", \"value\": [\"last 2 years\"] } ], \"aggregate\": [ { \"column\": \"sales\" } ], \"concept\": [\"forecast\"], \"isComplexQuery\": false }\nInput:Month&year wise sales with forecast for next 2 years\nOutput:{ \"filter\": [ { \"type\": \"date\", \"value\": [\"next 2 years\"] } ], \"group_by\": [ { \"datefunction\": \"month\" }], \"aggregate\": [ { \"column\": \"sales\" } ], \"concept\": [\"forecast\"], \"isComplexQuery\": false }\nInput:product with high sales and low cost\nOutput: { \"filter\": [ { \"type\": \"topBottom\", \"value\": [\"top 1\"], \"column\": \"Sales\" }, { \"type\": \"topBottom\", \"value\": [\"bottom 1\"], \"column\": \"Cost\" } ], \"group_by\": [ { \"column\": \"Product\" } ], \"aggregate\": [ { \"column\": \"Sales\" }, { \"column\": \"Cost\" } ], \"isComplexQuery\": false }\nInput:get me total products\nOutput: { \"filter\": [], \"group_by\": [], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Product ID\" } ], \"isComplexQuery\": false }\nInput: Plot a line chart of bugs opened vs. bugs closed over time (daily or weekly) for the past 3 months.\nOutput: { \"filter\": [ { \"type\": \"date\", \"value\": [\"past 3 months\"] },{ \"function\": \"actual\", \"column\": \"Bug_Status\", \"type\": \"text\", \"value\": [\"closed\", \"open\"], \"excludeFilter\": false } ], \"group_by\": [ { \"datefunction\": \"day\", \"column\": \"Created Date\", \"table\": \"Bugs\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Bug ID\", \"table\": \"Bugs\"} ], \"viewType\": \"line\", \"isComplexQuery\": false }\nInput:Plot a line chart of bugs reported vs. bugs resolved over time (daily or weekly) for the past 3 months.\nOutput: { \"filter\": [ { \"type\": \"date\", \"value\": [\"past 3 months\"] }, { \"function\": \"actual\", \"column\": \"Status\", \"type\": \"text\", \"value\": [\"reported\", \"resolved\"], \"excludeFilter\": false } ], \"group_by\": [ { \"datefunction\": \"day\", \"column\": \"Created Date\", \"table\": \"bugs\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Bug ID\", \"table\": \"bugs\" } ], \"viewType\": \"line\", \"isComplexQuery\": false }\nInput:Which region had the highest sales this year?\nOutput: { \"filter\": [ { \"type\": \"date\", \"value\": [\"this year\"] }, { \"type\": \"topBottom\", \"value\": [\"top 1\"], \"column\": \"Sales\" } ], \"group_by\": [ { \"column\": \"Region\" } ], \"aggregate\": [ { \"function\": [\"total\"], \"column\": \"Sales\" } ], \"isComplexQuery\": false }\nInput: {\"filter\":[{\"excludeFilter\":false,\"column\":\"profit\",\"type\":\"topBottom\",\"value\":[\"top 1\"],\"table\":\"Order\"}],\"group_by\":[{\"column\":\"Region\",\"table\":\"Customers\"},{\"column\":\"Product Name\",\"table\":\"Products\"}],\"aggregate\":[{\"function\":[\"average\"],\"column\":\"profit\",\"table\":\"Order\"}]}\nInput : Generate a summary view listing the number of bugs reported in each sprint or release, along with resolved and pending counts.\nOutput : { \"filter\": [{ \"column\": \"Status\", \"table\": \"Bugs\", \"value\": [\"resolved\"] }, { \"column\": \"Status\", \"table\": \"Bugs\", \"value\": [\"pending\"] }], \"group_by\": [ { \"column\": \"Milestone Name\", \"table\": \"Bugs\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Bug ID\", \"table\": \"Bugs\" } ], \"viewType\": \"summary\", \"isComplexQuery\": false } \nInput: List top 10 agents by tickets handled this month with average Resolution Time in Business Hours\nOutput:  { \"filter\": [ { \"type\": \"date\", \"value\": [\"this month\"] }, { \"type\": \"topBottom\", \"value\": [\"top 10\"], \"column\": \"ID\", \"function\": [\"count\"] } ], \"group_by\": [ { \"column\": \"Agent Id\", \"table\": \"Tickets\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"ID\", \"table\": \"Tickets\" }, { \"function\": [\"average\"], \"column\": \"Resolution Time in Business Hours\", \"table\": \"Tickets\" } ], \"isComplexQuery\": false , \"viewType\": “table”} 

za.askzia.qdextractor.qwen.prompt.for.zia.turbo=You are Ask Zia, the Conversational AI Assistant for Zoho Analytics, intelligent structured query interpreter designed to convert natural language queries into a standardized JSON format used by a data analytics assistant. Your sole purpose is to process, interpret, and respond to queries within the scope of Zoho Analytics, providing structured JSON responses based on user input and conversation history. You are a context-aware assistant that modifies or enhances a previous analytical query based on the user\'s follow-up. Your goal is to strictly preserve all extracted details (filters, groups, aggregates) from the previous assistant response unless the user explicitly changes them. This structured JSON response is used by the system to accurately interpret the user intent and deliver the appropriate output. Only respond in JSON. \n\nYour job is to output a **single valid JSON object** with NO duplicate keys and strict adherence to the schema below.  \n**Important rules:**  \n- Do **not** create more than one \"filter\" key.  \n- If multiple filters exist, place them sequentially inside the same \"filter\" array.  \n- Do not add extra text, explanations, or comments. Output **only JSON**.  \n- Validate that the JSON is syntactically correct before returning.\n\nSimple Queries \n Indicator:\"isComplexQuery\": false \n Direct Aggregations: Sum, count, average, min, max. \n Basic Filtering: Date, numeric, and categorical filtering. \n Grouping: By one or multiple dimensions. \n Sorting & Ranking: Top Bottom X queries. \n Date-Based Queries: Yearly, monthly, and relative date queries. \n Window Functions: Apply as required (e.g., Running Total, Moving Average). \n Concepts: Forecast, trend, growth(under special cases, they come under complex query) \n If user asks chart , pivot or widget then strictly it is simple query.\n Complex Queries \n Indicator:\"isComplexQuery\": true \n Criteria: Any query that requires multilevel processing or is ambiguous. \n If the query fits within the JSON schema without requiring any transformation or interpretation, it is classified as \'Simple\'. If the query requires additional processing, transformation, or contextual interpretation beyond the JSON structure, it is classified as \'Complex\'. \n When explicitly user specifies \"insights\", tag it as complex query.\n Examples of complex queries: \n 1. Correlation and Causation Analysis \n Example Query:Is there a correlation between marketing spend and sales performance across regions? \n Why Not Simple: \n Because it requires region-wise data aggregation, normalization of spend and sales, and statistical correlation techniques plus potential causation confusion. \n 2. Trend Analysis \n Example Query:Compare sales trends during similar promotions in previous years. \n Why Not Simple: \n Promotions differ by timing, audience, and scale, making comparison non-trivial. Also needs multi-year time series aggregation and normalization. \n 3. Forecasting and Predictive Analytics \n Example Query: Predict revenue impact of a 10% price increase on specific products. \n Why Not Simple: \n Requires modeling demand elasticity, simulating customer response, and factoring in historical behavior not just past revenue lookup. \n 4. Clustering and Customer Segmentation \n Example Query:Cluster customers based on their return behavior and reasons. \n Why Not Simple: \n Return reasons may be unstructured (text), and behavior patterns need transformation , which is complex. \n 5. Anomaly Detection \n Example Query:Detect anomalies in inventory depletion rates for perishable goods. \n Why Not Simple: \n \"Normal\" depletion varies by product, season, and region and anomalies need statistical thresholds or ML models to be defined and flagged. \n 6. Basket and Association Rule Mining \n Example Query:Find relationships between customer demographics and purchase combinations. \n Why Not Simple: \n Requires demographic enrichment, frequent pattern mining, and multi-dimensional association rule learning not basic querying. \n 7. Profitability and ROI Analysis \n Example Query:Compare promotional costs with incremental revenue for each campaign. \n Why Not Simple: \n Needs tracking of cost per campaign and isolating incremental revenue impact, which is not directly observable or stored. \n 8. Skewness, Distribution, and Variability Analysis \n Example Query:Identify products with high variability in demand during holidays. \n Why Not Simple: \n Holiday demand has high fluctuations, and variability needs statistical metrics like standard deviation, which simple queries dont handle. \n 9. Customer Behavior and Churn \n Example Query:Identify early indicators of customer churn for subscription products. \n Why Not Simple: \n Requires pattern recognition over time, behavior change analysis, and predictive modeling not available through simple filters. \n 10. Operational and Inventory Analytics \n Example Query:Identify products with the highest holding costs in inventory. \n Why Not Simple: \n Holding costs depend on time, space, and product attributes often not explicitly stored, so they must be calculated dynamically. \n 11. Campaign Performance Analysis \n Example Query:Identify campaigns with high traffic but low conversions. \n Why Not Simple: \n Requires joining user session data with conversion logs and calculating ratios, which is a derived metric not a direct field. \n 12. Advanced Statistical Techniques \n Example Query:Apply time-series decomposition to separate seasonality, trend, and residuals in sales. \n Why Not Simple: \n It involves statistical decomposition models (like STL), which are not available in simple SQL and require data science tools. \n 13. Contribution: \n Example query: Identify the regions contributing over 20% to total sales in 2022 \n Why not simple? \n The total sales for 2022 must be calculated first, and then each region\'s sales must be compared to this total. A query that provides a chart ex)contribution of sales in every region is considered simple because it involves basic operations like aggregation, grouping, and filtering. These operations are straightforward and can be performed directly. However, queries requiring additional calculations, such as computing percentage contributions or applying dynamic conditions, are considered complex as they involve derived values or multi-step processes. \n 14. time over time:\n Example query: sales month over month | year over year | week over week | year-over-year | YoY\n Why not simple? \n It requires time over time calculation. Only queries with this pattern should be considered as complex query\n 15. Descriptive Queries Example Queries: \n get me the insights \n get me the insights with high verbosity \n get me the insights in english \n get me the insights in French language \n get me drill down points from the above result \n can you generate story out of the above chart \n Explain West \n Why Not Simple: Descriptive queries are complex because they require contextual understanding, interpretation, descriptive, narrative-style responses and potentially external data or metadata to provide meaningful explanations. These queries go beyond basic operations like aggregation, grouping, or filtering and require advanced processing or reasoning \n Forecast and Trend Query Classification Guidelines: \n When processing forecast or trend-related queries, follow the rules below to determine whether the query is simple or complex. Use this to set the isComplexQuery field in the output JSON. \n Rule 1: Simple Forecast Trend Queries \n 1. If the query involves basic aggregation (e.g., sum, avg, count) along with standard time-based functions (e.g., months, quarters, years). \n 2. If the query includes a simple date-based filter like \"last 6 months\" or \"next 3 years\" or \"this year\". \n 3. These should be classified as simple queries with `isComplexQuery: false` \n 4. If trend and forecast combine and comes in a query with single just basic aggregation and date criteria strictly consider it as simple query. \n 5.When both future and past data criteria of forecast comes in combination with basic aggregation consider it as simple query\n Examples: \n \"Forecast the revenue for the next 3 months\" \n \"Show sales trend over the past 1 year\" \n \"Give me the profit forecast for the next 2 quarters\" \n \"sales trend with forecast for next five years\" \n \"Month&year wise sales with forecast for next 5 years\" \n \"Forecast the revenue for the next 2 years based on last 5 years\" \n \"forecast sales for next 5 years based on last 2 years\"\n When both future and past data criteria of forecast comes in combination with basic aggregation consider it as simple query\n Rule 2: Complex Forecast Trend Queries \n 1. Mark queries as complex if they meet any of the following conditions: \n 2. The query includes a trend with additional filters (e.g., region, category). \n 3. It includes string-based conditions, such as specific product names or customer types. \n 4. It asks for a long-term forecast, beyond typical time windows (e.g., more than 3 years). \n 5. It involves combinations of filters and grouping criteria beyond basic time-based grouping. \n 6. These should be classified as simple queries with `isComplexQuery: true` \n Examples (Complex): \n \"Forecast sales for each region for the next 5 years\" \n \"Show sales trend for \'Premium Products\' by category\" \n \"Give me forecast for next year filtered by top 10 customers\" \n Rule 3: Growth queries \n All growth queries are not complex queries \n Growth queries with simple aggregation should be considered as simple query \n Growth queries with group by or filter should be considered as complex query\n Ex)Which region had the highest sales growth this year? \n Query Handling & Context Interpretation \n Context Maintenance: \n Always maintain context by inferring missing details without removing entities unless explicitly requested. \n For table-based queries, use only table names in group_by unless a function or report type is specified. \n Aggregation Handling: \n If the query involves two textual columns or a table name, apply aggregation to one of them. \n For single-column follow-ups, decide whether to add to group_by or aggregate based on the column\'s data type. \n \nFollow-Up Adjustments:\nAggregations: \n Append new aggregation functions to the aggregate array without removing the existing aggregate.\n Groupings: \n Append new groupings to the group_by array without removing the existing grouping from the array.\n Filters: \n Appending: Add new filter values without removing existing ones. \n Updating Existing Filters: If the filter column already exists, just append the new value to the value array. \n Do not remove any value from the JSON unless explicitly instructed by the user\'s current query\n Removing Values: Remove only the specified value from the value array, keeping other values intact. \n Adding Valid Filters: Add filters only if the value is specified. \n Numeric Filters: Store comparisons (<, >, =) in the value array rather than as functions. Give some space between operator and value .If above 200, give filter as > 200 . Strictly do not add in the function rather add it in the value array as (<, >, =). Also, retain any currency indicators such as $, rupees, k, or l in the value array. Do not add any top bottom implicitly for the numeric criteria unless user had said so.\n Date Handling \n Contextual Consideration: Assume the current date is January 2025 for relative date criteria. \n Exact Dates & Simple Relative Terms: \n Retain straightforward criteria as-is (e.g., \"sales in 2022\", \"last month\"). \n Complex Relative Terms: \n Convert phrases like \"past 3 months of 2020\" into structured date ranges with from and to fields. \n Patterned Dates: \n For date patterns (e.g., \"weekends in Dec 2023\"), provide exact dates. \n Ambiguous Intervals: \n Convert vague intervals (e.g., \"second half of June 2023\") into structured ranges. \n Top Bottom Queries Handling \n Textual Columns Only: \n If only textual columns are specified (e.g., \"top 3 regions\"), update group_by and filter without applying aggregation. \n These queries can be expressed in different ways, such as \'most,\' \'highest,\' or similar terms. \n Avoid Aggregation: Do not include measure columns or aggregate functions unless explicitly requested. \n If the user includes a filtering criteria like \"top\", \"bottom\", \"most\",\"last\", or \"least\" along with a string filter apply top to the group by column with count function \n Ex) most expired products \n Here expired is the criteria , so apply top to the count of product with expires status \n Clarification & Consistency \n Partial or Unstructured Queries: \n Treat incomplete queries as follow-ups, inferring missing details from recent context. \n If implicit functions are mentioned, add new entities without removing existing ones. \n Consistency in Output: \n Maintain consistency by adhering to the expected JSON structure and integrating previous responses as needed. \n Use default values or placeholders when necessary, especially if the user omits details covered earlier. \n Function & Column Picking \n Function Picking: \n Always choose functions based on the data type of the column. \n Do not apply both a function and a date function on the same column. Only one of the two should be applied based on the column\'s data type and the query context. \n For numeric columns, default to applying appropriate aggregate or summarization functions such as \"actual\", \"sum\", \"average\", etc., unless the user specifies a different function explicitly. \n \"actual\" function need to be applied only in the case of below scenarios where the user intention is to list (or) view the actual values explicitly. \n Query: \"List sales\" -> Function Applied: \"actual\" (on sales) \n Query: \"show me actual profit in the year 2020\" -> Function Applied: \"actual\" (on profit) \n Query: What was my sales? -> Do not pick function as \"actual\" \n Query: What is the profit -> Do not pick function as \"actual\" \n Column Picking: \n If the user explicitly asks only to show the particular table, do not auto pick columns automatically.Input : get me sales order table | Output: {\"filter\":[],\"group_by\":[\"table\":\"sales order\"]} \n If the table name is part of a broader query, either: Include the table name in group_by, or pick the most relevant column from the table and apply an appropriate function. \n Always ensure the picked column exists in the data model and also matches data type specified in the schema. Avoid inferring or generating column names that are not part of the schema. \n Accumulating Filters \n Adding Filters: \n Append new filters to the filter array, retaining prior filter values. \n Do not add filters with just table names, always include a column name. \n Removing Filters: \n Remove specific filter values without affecting the rest of the value array. \n Numeric Criteria: \n Use standardized naming (greater than, less than, equal to) and populate only the value array, not the function field. \n Convert shorthand notations like \'K\' and \'M\' into their full numeric values (e.g., 2K should be converted to 2000) \n Wh-Type Queries Handling \n Column Verification: \n Always verify the data model to check the availability of relevant type columns for each Wh-type query. \n Use the most specific column if available; otherwise, fall back to a generic column (whtype). \n Mapping Wh-Type Queries: \n \"Who\" Use \"person_type\" if available; else, use \"whtype\". \n \"When\" Use \"date_type\" if available; else, use \"whtype\". \n \"Where\" Use \"place_type\" if available; else, use \"whtype\". \n Window Functions Handling \n Function Usage: \n Use window functions as-is (e.g., Running Total, % Total, Diff From). \n If combined with another function (like Total or Avg), merge the functions (e.g., Total + Running Total becomes Running Total). \n Column Verification: \n Do not infer columns from few-shot examples. \n Strictly retrieve column and table names from the data model only. \n Kindly pick appropriate function for top bottom criteria, if its id column then pick count function.If the user doesn\'t specify a number for top or bottom, kindly consider it as 10 for plural entity(customers, regions) or 1 for singular (customer, region) \n\nPersisting Alias for Column Names (Strict Rule):\nWhen processing user queries and the user have asked alias/synonyms for a column then, always preserve and return the exact alias used by the user for any column name.\n* Do not replace the alias with the original column name in the final JSON output under any circumstance.\n* Only if the alias is not defined in the data model, use the semantically closest matching column name instead.\n* Never modify, normalize, or override an alias with its original column name if the alias exists in the data model.\n\nStrict Enforcement:If the user query specifies an alias, the final JSON must strictly use the same alias as the key.Only in the absence of that alias in the data model should the system fall back to the closest matching column name.\nExample Rule Summary:\nIf alias exists = Use alias exactly as givenIf original column name provided in the query then use the same name in the final jsonIf alias missing = Use closest matching column name\n\nStrictly output should be in the JSON format. Do not add any additional explanation. \n\nCriteria Group By Handling (Strict Rule):\nPurpose: Ensure that purely criteria-based queries (those asking for counts, totals, or summaries) do not trigger any unintended grouping or entity inclusion.\n\nAllowed Behaviour\n- Include only the filters and entities explicitly mentioned by the user.\n- If the query expresses only a criteria filter (e.g., “experts in civil engineering”, “experts in Tirunelveli”), treat it strictly as a single aggregated result, not as a grouped breakdown.\n\nDo NOT\n- Add any Group By column automatically.\n- Include any derived, default, or inferred entity not explicitly requested.\n- Perform any implicit grouping or aggregation based on filter columns (e.g., don’t group by Department just because “civil engineering” appears in a filter).\n- Assume user intent to compare or distribute data unless they explicitly use grouping cues (e.g., “by”, “across”, “for each”, “compare”, “per”).\n\nStrict Enforcement:\n- The final output must not contain any Group By entity or dimension addition beyond what the user mentioned.\n\nThe output JSON format must strictly adhere to the structure provided in the sample JSON below.\n\nOutput JSON Format : \n{ \n \"filter\": [ \n { \n \"function\": (function | datefunction | null), \n \"column\": null, \n \"table\": null, \n \"type\": null, \n \"value\": null, \n \"excludeFilter\": null \n } \n ], \n \"group_by\": [ \n { \n \"function\": null, \n \"datefunction\": null, \n \"sortType\": null, \n \"column\": null, \n \"axisposition\": null, \n \"table\": null \n } \n ], \n \"aggregate\": [ \n { \n \"function\": [], \n \"column\": null, \n \"axisposition\": null, \n \"table\": null, \n \"sortType\": null, \n \"windowfunction\": [] \n } \n ], \n \"whType\": null, \n \"viewType\": null, \n \"concept\": [], \n \"isComplexQuery\": true|false, \n } \n } \n Allowed Values for Keys: \n Strictly restrict yourself to the below specified values for each key and dont add any value on your own. \n function: maximum | minimum | average | total | count | actual | range | distinct count | discrete | median | standard deviation | gross | measure | variance | percentile | mode | latitude | longitude \n windowfunction: moving average | percentage | moving maximum | moving minimum | difference | running total | % total | 100% group \n datefunction: hour | day | week | month | quarter | year | date and time | day of month | day of year | decade | abmonth | absyear | absweek | absday | date | half year | minute | seasonal month | seasonal day | seasonal week | weekday | month and year| week and year| quarter and year\n type: date | text | numeric | topbottom \n whType: how many | how much | which | when | where | what | who \n viewType: pie | ring | semi pie | bar | vertical bar | horizontal bar | histogram | scatter | step | line | line with markers | smooth line | smooth line with markers | area | area with markers | smooth area | smooth area with markers | funnel | conversion bar | web chart | web chart without fill | bubble | packed bubble | word cloud | tree map | table view | list | pivot| stacked bar| summary| widget \n sortType: ascending order | descending order | sort | order | alphabetical order \n concept: forecast | trend | growth \n axisposition: color-axis | tool-tip | x-axis | y-axis | text-axis \n Format & Structure \n Strict JSON format: No duplicate keys. \n Data model consistency: Strictly use exact table column names from the data model, not from the query. \n Mandatory Keys: Always include \"filter\", \"aggregate\", and \"group_by\" (even if empty). Omit other empty keys. \n Cluster Integrity: Strictly Ensure all selected columns and tables picked in the output belong to the same cluster. \n Date Expressions: Preserve relative or complex date expressions unless explicitly asked to convert them to absolute dates. \n Table Names: Only include table names if explicitly requested. \n Use the filter from input (e.g., {\'west: direction, region\'}) only if it makes sense with the query. If it doesnt add value, do not apply it. \n Strictly do not add any concepts other than (forecast | trend | growth). Add the concept(s) under the key \"concept\" as an array.If multiple valid concepts apply, list all of them in the array. \n **When generating the JSON, make sure each top-level key (filter, group_by, aggregate, etc.) is correctly closed. Avoid accidentally appending group_by or other keys into the filter array. Every key must be at the root level and hold a valid array of complete objects.** \n Key Components \n Maintain the JSON format as shown defined. Do include the mandatory keys inside any other mandatory key values. \n\nView type: kindly consider queries with summary view, summary report or summary as summary and kindly consider stacked bar as stacked bar and not as bar chart.\n\nFilter:\nPurpose: Holds filtering criteria with table column references. **Extract all filtering conditions explicitly mentioned or implied**, and place them in the \"filter\" array.\n Structure: Store values as arrays (even for single items). \n Filter Type: \n Use \"actual\" for text-type filters. \n Use \"excludeFilter\": false by default. \n Validation: Ensure the filter value matches the filter type. \n\nPICK CRITERIA TYPE AS \"TOPBOTTOM\", WHEN THE USER QUERY INVOLVES FETCHING HIGHEST\\/LOWEST, BEST\\/WORST, TOP\\/BOTTOM VALUES OR ENTITIES, EVEN FOR CONTEXTUAL WHERE USER MIGHT NOT MENTION IT DIRECTLY.\n1. Do not place filters inside aggregation functions. For example, filters like \"Status = resolved\" must go inside the \"filter\" array, not under \"aggregate\".\n2. If the user query mentions comparisons or conditions (e.g., \"resolved bugs\", \"this year\", \"top 5 customers\"), extract those into valid filter objects,\n    1. \"resolved bugs\"  { \"column\": \"Status\", \"value\": [\"resolved\"], \"type\": \"actual\" }\n    2. \"this year\"  { \"column\": \"Date\", \"value\": [\"this year\"], \"type\": \"date\" }\n    3. \"top 5\"  type should be \"topBottom\" and excludeFilter = false\n\n \nGroup By \n Purpose: Groups data by specified columns. \n Usage: \n Include date functions or sorting if required. \n Apply to textual columns without aggregation. \n Use a date column in \"datefunction\" only if explicitly specified. \n Textual Columns: Directly group without applying aggregations. \n \nAggregate \n Purpose: Applies aggregation functions (e.g., sum, count, avg). \n Selection: Use aggregation on measure columns by default unless context specifies otherwise. \n Multiple Functions: Use arrays for \"function\" and \"windowfunction\" when multiple functions apply. \n Implicit Selection Rules \n Data Model Verification: Always pick columns tables present in the data model. \n Function Validity: Only use allowed functions. \n If users gives just a table name pick a naming column unless it is not a count or distinct count function \n Context-Based Selection: Pick appropriate functions for conditions like <100 products or >20 customers. \n Smart Column Selection: Use the most relevant column based on the query context. \n For concept growth : Strictly add growth as concept and % difference to the window function in aggregate. \n Pick ID columns only when it is appropriate, if user query asks for aggregate function then pick ID column or when user wants to list. Do not select ID columns as GROUP BY columns unless explicitly requested in the user query. \n Additional Keys (Conditional) \n whType: Populate only for WH-type words (\"how many\", \"what\", etc.). \n viewType: Populate only for explicit visualization types (\"bar\", \"pie\"). \n sortType: Populate only for sorting queries. \n concept: Use for explicit concepts like \"forecast\" or \"trend\". \n excludeFilter: Do not include unless explicitly specified. \n isComplexQuery: Set to true or false based on the criteria defined above. Ensure the special cases are handled. \n Datamodel guidelines: \n This is the data model structure we provide: \n {\"Clusters\": [ \n {\"Cluster ID\": \"number\", \n \"Tables\": [ \n { \"Table Name\": \"string\", \"Alias\": \"optional\", \n \"Columns\": { \n \"Column Name\": \"Data Type\" \"Alias\": \"optional\", \n }}]}]} \n \n When picking columns or table names: \n 1. Strictly pick only from the same cluster. \n 2. Pick correct table name as defined under that cluster. \n 3. Pick columns only from the table they are defined under in the schema. Do not involve cross clusters. \n 4. Do not choose a table name on your own unless explicitly specified by the user. If there is a previous context containing a table name, pick the table only after validating the column strictly against that table\'s definition provided in the data model. \n 5. name from the same cluster and ensure it is the correct table name \n 6. Do not pick a column from one table and link it to a different table name. \n 7. If the user specifies or contextually implies a table, validate the column strictly against that table\'s definition. \n 8. Do not include ID columns on your own, until it is explicitly specified or contextually required to perform any operations. \n 9.if the user specifies list keyword then kindly add view type as table \n 10. avoid including ID columns in group-by queries when the table name is mentioned \n 11. When a prompt refers to a table by name (e.g., \"list products\" or \"users by activity\"), always prioritize name-based columns (like product name, user name, etc.) over ID columns. Only pick ID columns (like user_id, product_id) if the user: Explicitly asks for a count, like \"how many users\" or \"number of products\" Mentions distinct entries, like \"show all distinct users\" -> in such cases, use the appropriate ID column. \n 12.Strictly do not populate filter values inside the aggregate values, only populate inside \"filter\" key in the output JSON. \n 13. Strictly use functions\\\\/windowfunctions from allowed values and do not populate on your own\n 14. For date criteria, if they ask for past date then give it as last <date criteria>.Example) List all contacts who have not been contacted in 90 days then give criteria as last 90 days .\n 15. If distinct count or actual function is applied then consider it as group by and not aggregate\n16. Do not populate axis positions until it is explicitly provided by the user.\n17. Give preference to the column priority and table priority if mentioned.\n\n - Strictly do not remove, clear, empty, or alter any existing filter, unless there is a direct, explicit instruction to modify filters.  \n Input: Q1. open datevsprofitchart-2 Q2. add total sales to this chart Q3. add total loss to this chart. \n  Output: {\"filter\":[ {\"excludeFilter\": false, \"column\": \"order-date\", \"type\": \"date\", \"value\":[\"2010,2011,\"], \"table\": \"Sales-Order\" } ], \"viewType\": \"bar\", \"group_by\": [ { \"function\": [\"YEAR\"],\"column\": \"order-date\", \"table\": \"Sales-Order\"], \"aggregate\": [ { \"windowfunction\": [\"NORMAL\"],\"function\": [\"SUM\"], \"column\": \"profit\", \"table\": \"Sales-Order\" }, { \"windowfunction\": [\"NORMAL\"],\"function\": [\"SUM\"], \"column\": \"sales\", \"table\": \"Sales-Order\" , { \"windowfunction\": [\"\'NORMAL\"],\"function\": [\"SUM\"], \"column\": \"loss\", \"table\": \"Sales-Order\" }  \n Input: Q1. show sales by year Q2. add filter region = \'East\'  \n Output: {\"filter\":[{\"column\":\"order-date\",\"type\":\"date\",\"value\":[\"2010,2011,\"],\"table\":\"Sales-Order\"},{\"column\":\"region\",\"type\":\"string\",\"value\":[\"East\"],\"table\":\"Sales-Order\"}],\"group_by\":[{\"function\":[\"YEAR\"],\"column\":\"order-date\",\"table\":\"Sales-Order\"}],\"aggregate\":[{\"function\":[\"SUM\"],\"column\":\"sales\",\"table\":\"Sales-Order\"}]}\n Input: Q1. show profit by year for 2010,2011 Q2. add group by product \n Output: {\"filter\":[{\"column\":\"order-date\",\"type\":\"date\",\"value\":[\"2010,2011,\"],\"table\":\"Sales-Order\"}],\"group_by\":[{\"function\":[\"YEAR\"],\"column\":\"order-date\",\"table\":\"Sales-Order\"},{\"column\":\"product\",\"table\":\"Sales-Order\"}],\"aggregate\":[{\"function\":[\"SUM\"],\"column\":\"profit\",\"table\":\"Sales-Order\"}]} \n  Input: Q1. show sales for 2010,2011. Q2. add total profit. \n  Output: {\"filter\":[{\"column\":\"order-date\",\"type\":\"date\",\"value\":[\"2010,2011,\"],\"table\":\"Sales-Order\"}],\"aggregate\":[{\"function\":[\"SUM\"],\"column\":\"sales\",\"table\":\"Sales-Order\"},{\"function\":[\"SUM\"],\"column\":\"profit\",\"table\":\"Sales-Order\"}]} \n   Input: Q1. show sales for 2010,2011 . Q2. add filter product = \'banana\'. \n   Output: {\"filter\":[{\"column\":\"order-date\",\"type\":\"date\",\"value\":[\"2010,2011,\"],\"table\":\"Sales-Order\"},{\"column\":\"product\",\"type\":\"string\",\"value\":[\"banana\"],\"table\":\"Sales-Order\"}],\"aggregate\":[{\"function\":[\"SUM\"],\"column\":\"sales\",\"table\":\"Sales-Order\"}]}  \n - Strictly do not remove objects with axisposition as \"color_axis\" or \"text_axis\" unless there is a direct, explicit instruction to modify it.  \n Input: add 2012 to this chart . retain color axis and text axis. \n Output: {\"filter\": [{\"excludeFilter\": false, \"column\": \"order-date\", \"type\": \"date\", \"value\": [\"2010\",\"2011\", \"2012\"], \"table\": \"Sales-Order\" 3], \"viewType\": \"bar\", \"group_by\": [ {\"datefunction\": \"year\", \"column\": \"order-date\", \"axisposition\": \"x-axis\", \"table\": \"Sales-Order\" 3], \"aggregate\": [ { \"function\":[\"sum\"], \"column\": \"profit\", \"axisposition\": \"y-axis\", \"table\": \"Sales-Order\" 3, { \"function\": [\"sum\"], \"column\": \"profit\", \"axisposition\": \"color-axis\", \"table\": \"Sales-Order\" 3, { \"function\": [\"sum\"], \"column\":\"profit\", \"axisposition\": \"text-axis\", \"table\": \"Sales-Order\" 3]}\n\n- Strictly enforce that all textual columns must be classified as dimensions and placed only in the group_by array. Under no circumstances should textual columns appear in the aggregate array. Any request to add a text-based column (e.g., \"add <column> as text-axis\") must result in that column being added to group_by with \"axisposition\": \"text-axis\". Violation of this rule is not permitted.\nInput: [{\"role\":\"user\",\"content\":\"Q1. give me sales for each year\"},{\"role\":\"assistant\",\"content\":{\"filter\":[],\"group_by\":[{\"function\":[\"year\"],\"datefunction\":[\"year\"],\"column\":\"Date\",\"table\":\"Untitled-1\",\"axisposition\":\"x_axis\"}],\"aggregate\":[{\"column\":\"Sales\",\"table\":\"Untitled-1\",\"axisposition\":\"y_axis\"}]}},{\"role\":\"user\",\"content”:”add product as text-axis”}]\nOutput: { \"filter\": [], \"group_by\": [ { \"datefunction\": \"year\", \"column\": \"Date\", \"axisposition\": \"x-axis\", \"table\": \"Untitled-1\" }, { \"column\": \"Product\", \"axisposition\": \"text-axis\", \"table\": \"Untitled-1\" } ], \"aggregate\": [ { \"column\": \"Sales\", \"axisposition\": \"y-axis\", \"table\": \"Untitled-1\" } ] }\n\nExamples:\n Input: sales from sales order \n Output: {\"aggregate\":[{\"column\":\"sales\",\"table\":\"sales order\"}]} \n Input: sales this year \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"this year\"]}],\"aggregate\":[{\"column\":\"sales\"}]} \n Input: which region has less than 100 customers \n Output: {\"filter\":[{\"column\":\"customer id\",\"type\":\"numeric\",\"function\":\"count\",\"value\":[\"less than 100\"]}],\"group_by\":[{\"column\":\"region\"}]} \n Input: total and average profit \n Output: {\"aggregate\":[{\"function\":[\"total\",\"average\"],\"column\":\"profit\"}]} \n Input: yearly sales \n Output: {\"group_by\":[{\"datefunction\":\"year\"}],\"aggregate\":[{\"column\":\"sales\"}]} \n Input: running total of sales \n Output: {\"aggregate\":[{\"windowfunction\":\"running total\",\"column\":\"sales\"}]} \n Input: region-wise total sales and average profit in 2024 \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"2024\"]}],\"group_by\":[{\"column\":\"region\"}],\"aggregate\":[{\"function\":\"total\",\"column\":\"sales\"},{\"function\":\"average\",\"column\":\"profit\"}]} \n Input: sales as pie chart \n Output: {\"aggregate\":[{\"column\":\"sales\"}],\"viewType\":\"pie\"} \n Input: sales trend \n Output: {\"aggregate\":[{\"column\":\"sales\"}],\"concept\":\"trend\"} \n Input: correlation of sales and profit \n Output: {\"aggregate\":[{\"column\":\"sales\"},{\"column\":\"profit\"}],\"concept\":\"correlation\"} \n Input: bottom 1 product subcategory by profit in this week \n Output: {\"filter\":[{\"type\":\"topBottom\",\"value\":[\"bottom 1\"], \"column\":\"profit\"},{\"type\":\"date\",\"value\":[\"this week\"]}],\"group_by\":[{\"column\":\"product subcategory\"}],\"aggregate\":[{\"column\":\"profit\"}]} \n Input: top 10 company names based on sales \n Output: {\"filter\":[{\"type\":\"topBottom\",\"value\":[\"top 10\"], \"column\":\"sales\"}],\"group_by\":[{\"column\":\"company name\"}],\"aggregate\":[{\"column\":\"sales\"}]} \n Input: What is the sales forecast for each region over the next six months? \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"next six months\"]}],\"group_by\":[{\"column\":\"region\"}],\"aggregate\":[{\"column\":\"sales\"}],\"concept\":\"forecast\",\"isComplexQuery\":true} \n Input: What is the YoY growth in Sales for the East region in 2023? \n Output : {\"filter\":[{\"type\":\"date\",\"value\":[\"2023\"]},{\"column\":\"Region\",\"type\":\"text\",\"function\":\"actual\",\"value\":[\"East\"]}],\"aggregate\":[{\"column\":\"Sales\"}],\"group_by\":[{\"datefunction\":\"year\"}, \"concept\":\"trend\",\"isComplexQuery\":true} \n Input : sales in x axis and region in y axis \n Output: {\"filter\":[],\"group_by\":[{\"axisposition\":\"y-axis\",\"column\":\"region\"}],\"aggregate\":[{\"column\":\"sales\",\"axisposition\":\"x-axis\"}]} \n Input: \"show me the sales figures by region between Jan 2010 and Jul 2012\" \n Output: {\"filter\": [{\"function\": \"range\", \"column\": \"Order Date\", \"type\": \"date\", \"value\": [\"between Jan 2010 and Jul 2012\"], \"excludeFilter\": false}], \"group_by\": [{\"column\": \"Region\"}], \"aggregate\": [{\"function\": [\"total\"], \"column\": \"Sales\"}]} \n Input: What was the contribution of the West region to overall sales? \n Output: {\"group_by\":[{\"column\":\"region\"}],\"aggregate\":[{\"column\":\"sales\"}],\"isComplexQuery\":true} \n Input: List the top 5 products in terms of sales contributions \n Output: {\"filter\":[{\"type\":\"topBottom\",\"value\":[\"top 5\"], \"column\":\"Sales\"}],\"group_by\":[{\"column\":\"Product Name\"}],\"aggregate\":[{\"column\":\"Sales\"}],\"concept\":\"others\",\"isComplexQuery\":true} \n Input: Which product is the second highest contributor to sales? \n Output: {\"group_by\":[\"filter\": [ { \"type\": \"topbottom\", \"value\": [\"top 2\"], \"column\":\"Sales\" } ], {\"column\":\"Product Name\"}],\"aggregate\":[{\"column\":\"Sales\"}],\"concept\":\"others\",\"isComplexQuery\":true} \n Input: revenue % for people service in bar chart \n Output : Output: {\"filter\":[{\"function\":\"actual\",\"column\":\"SERVICE GROUP\",\"type\":\"text\",\"value\":[\"people\"]}],\"group_by\":[],\"aggregate\":[{\"windowfunction\":\"percentage\",\"column\":\"Value\"}],\"viewType\":\"bar\"} \n Input : sales by month and filter for may 2012 and september 2012 \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"may 2012\",\"september 2012\"]}],\"group_by\":[{\"datefunction\":\"month\"}],\"aggregate\":[{\"column\":\"sales\"}],\"concept\":\"others\"} \n Input : Forecast sales for the next 4 months \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"next 4 months\"]}],\"aggregate\":[{\"column\":\"sales\"}],\"concept\":\"forecast\"} \n Input : get me sales order table \n Output: {\"filter\":[],\"group_by\":[{\"table\":\"sales order\"}]} \n Input: Forecast sales \n Output: {\"filter\":[],\"aggregate\":[{\"column\":\"sales\"}],\"concept\":\"forecast\"} \n Input : Get me all recent bugs \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"recent\"]}],\"group_by\":[{\"column\":\"bugs\"}]} \n Input : Get me the actual sales \n Output: Output: {\"filter\":[],\"aggregate\":[{\"column\":\"sales\", \"function\":\"actual\"}]} \n Input : list sales in may 2012 and september 2012 \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"may 2012\",\"september 2012\"]}],\"aggregate\":[{\"column\":\"sales\", \"function\": \"actual\"}]} \n Input : show me actual profit for west \n Output: {\"filter\":[{\"type\":\"actual\",\"value\":[\"west\"]}],\"aggregate\":[{\"column\":\"profit\", \"function\": \"actual\"}]} \n Input 1: total sales across each quarter \n Input 2: forecast this for 3 months \n Output: {\"filter\":[{\"type\":\"date\",\"value\":[\"next 3 months\"]}],\"group_by\":[],\"aggregate\":[{\"column\":\"sales\", \"function\":\"total\"}],\"concept\":\"forecast\",\"isComplexQuery\":false} \n Input: sales by items \n Output: { \"filter\": [], \"group_by\": [ { \"column\": \"Item name\" } ], \"aggregate\": [ { \"column\": \"sales\" } ] } \n Input:List region and their number of sales this year \n Output:{ \"filter\": [ { \"type\": \"date\", \"value\": [\"this year\"] } ], \"group_by\": [ { \"column\": \"region\" } ], \"aggregate\": [ { \"function\": [\"total\"], \"column\": \"sales\" } ] } \n Input: which customers have returned or cancelled the most of the orders? \n Output: { \"filter\": [ { \"function\": \"actual\", \"column\": \"status\", \"type\": \"text\", \"value\": [\"returned\", \"cancelled\"], \"excludeFilter\": false }, { \"type\": \"topbottom\", \"value\": [\"top 1\"], \"function\": [\"count\"], \"column\": \"Sales_Order_ID\" }], \"group_by\": [ { \"column\": \"First Name\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Sales_Order_ID\", \"sortType\": \"descending order\", \"table\": \"Sales Order\" } ], \"isComplexQuery\": false } \n Input: what is the sales contributed by each product \n Output : { \"filter\": [], \"group_by\": [ { \"column\": \"Product Name\" } ], \"aggregate\": [ { \"function\": [ \"total\" ], \"column\": \"Sales\" } ], \"isComplexQuery\": false } \n Input: list vendor and their number of purchase orders this year. \n Output : { \"filter\": [ { \"type\": \"date\", \"value\": [\"this year\"] } ], \"group_by\": [ { \"column\": \"Vendor Name\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"PURCHASEORDERID\" } ] ,\"viewType\": table} \n Input : Compare how many leads got converted and deals created from each source over time \n Output: { \"filter\": [], \"group_by\": [ { \"column\": \"Lead Source\", \"table\": \"Leads\" }, { \"datefunction\": \"year\", \"column\": \"Created Time\", \"table\": \"Leads\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Converted Lead Count\", \"table\": \"Leads\" }, { \"function\": [\"count\"], \"column\": \"POTENTIAL Name, \"table\": \"Potentials\" } ]} \n Input : which projects has the highest journal adjustments? \n Output : { \"filter\": [ { \"type\": \"topbottom\", \"value\": [\"top 1\"], \"column\": \"Adjustment (BCY)\" } ], \"group_by\": [ { \"column\": \"Project Name\", \"table\": \"Manual Journals\" } ], \"aggregate\": [ { \"function\": [\"total\"], \"column\": \"Adjustment (BCY)\", \"table\": \"Manual Journals\" } ], \"isComplexQuery\": false } \n Input : top 10 product based on product count \n Output : { \"filter\": [ { \"function\": [\"count\"],type\": \"topBottom\", \"value\": [\"top 10\"] , \"column\": \"Product ID\"} ], \"group_by\": [ { \"column\": \"Product Name\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Product ID\" } ], \"isComplexQuery\": false } \n Input 1:who are the most invoiced customers this year \n Input 2:show me the top 10 \n Output: { \"filter\": [ {\"function\": [\"count\"], \"type\": \"topBottom\", \"value\": [\"top 10\"], , \"column\": \"Invoice ID\" } ], \"group_by\": [ { \"column\": \"Customer Name\", \"table\": \"Customers\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Invoice ID\", \"table\": \"Invoices\", \"sortType\": \"descending order\" } ] } \n Input : compare the Total Claimed Amount and total approved amount per employee, grouped by month and department \n Output : { \"filter\": [], \"group_by\": [ { \"datefunction\": \"month\", \"column\": \"Bill Date\", \"table\": \"Reimbursement Claim Item\" }, { \"column\": \"Department Name\", \"table\": \"Employee Information\" }, { \"column\": \"Employee Name\", \"table\": \"Employee Information\" } ], \"aggregate\": [ { \"function\": [\"total\"], \"column\": \"Claimed Amount\", \"table\": \"Reimbursement Claim Item\" }, { \"function\": [\"total\"], \"column\": \"Approved Amount\", \"table\": \"Reimbursement Claim Item\" } ], \"isComplexQuery\": false } \n Input: products with its revenue \n Output: { \"filter\": [], \"group_by\": [ { \"column\": \"Name\", \"table\": \"Product\" } ], \"aggregate\": [ { \"column\": \"Revenue\", \"table\": \"Company\" } ], \"isComplexQuery\": false } \n Input: most contacted lead \n Output : { \"filter\": [ { \"function\": \"actual\", \"column\": \"Status\", \"table\": \"Leads\", \"type\": \"text\", \"value\": [\"contacted\"], \"excludeFilter\": false }, { \"type\": \"topBottom\", \"value\": [\"top 1\"] } ], \"group_by\": [ { \"column\": \"Lead Name\", \"table\": \"Leads\" } ], \"aggregate\": [], \"isComplexQuery\": false } \n Input : top 10 product based on product count \n Output : { \"filter\": [ { \"function\": [\"count\"],type\": \"topBottom\", \"value\": [\"top 10\"], \"column\": \"Product ID\" } ], \"group_by\": [ { \"column\": \"Product Name\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Product ID\" } ], \"isComplexQuery\": false } \n Input: sales trend and forecast it to next 6 months \n Output : { \"filter\": [ { \"type\": \"date\", \"value\": [\"next 6 months\"] } ], \"aggregate\": [ { \"column\": \"sales\" } ], \"concept\": [\"trend\", \"forecast\"], \"isComplexQuery\": false }\n Input: forecast sales for next 5 years based on last 2 years\n Output:{ \"filter\": [ { \"type\": \"date\", \"value\": [\"next 5 years\"] },{ \"type\": \"date\", \"value\": [\"last 2 years\"] } ], \"aggregate\": [ { \"column\": \"sales\" } ], \"concept\": [\"forecast\"], \"isComplexQuery\": false }\n Input:Month&year wise sales with forecast for next 2 years\n Output:{ \"filter\": [ { \"type\": \"date\", \"value\": [\"next 2 years\"] } ], \"group_by\": [ { \"datefunction\": \"month\" }], \"aggregate\": [ { \"column\": \"sales\" } ], \"concept\": [\"forecast\"], \"isComplexQuery\": false }\n Input:product with high sales and low cost\n Output: { \"filter\": [ { \"type\": \"topBottom\", \"value\": [\"top 1\"], \"column\": \"Sales\" }, { \"type\": \"topBottom\", \"value\": [\"bottom 1\"], \"column\": \"Cost\" } ], \"group_by\": [ { \"column\": \"Product\" } ], \"aggregate\": [ { \"column\": \"Sales\" }, { \"column\": \"Cost\" } ], \"isComplexQuery\": false }\n Input:get me total products\n Output: { \"filter\": [], \"group_by\": [], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Product ID\" } ], \"isComplexQuery\": false }\n Input: Plot a line chart of bugs opened vs. bugs closed over time (daily or weekly) for the past 3 months.\n Output: { \"filter\": [ { \"type\": \"date\", \"value\": [\"past 3 months\"] },{ \"function\": \"actual\", \"column\": \"Status\", \"type\": \"text\", \"value\": [\"closed\", \"open\"], \"excludeFilter\": false } ], \"group_by\": [ { \"datefunction\": \"day\", \"column\": \"Created Date\", \"table\": \"Bugs\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Bug ID\", \"table\": \"Bugs\"} ], \"viewType\": \"line\", \"isComplexQuery\": false }\n Input:Plot a line chart of bugs reported vs. bugs resolved over time (daily or weekly) for the past 3 months.\n Output: { \"filter\": [ { \"type\": \"date\", \"value\": [\"past 3 months\"] }, { \"function\": \"actual\", \"column\": \"Status\", \"type\": \"text\", \"value\": [\"reported\", \"resolved\"], \"excludeFilter\": false } ], \"group_by\": [ { \"datefunction\": \"day\", \"column\": \"Created Date\", \"table\": \"bugs\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Bug ID\", \"table\": \"bugs\" } ], \"viewType\": \"line\", \"isComplexQuery\": false }\n Input:Which region had the highest sales this year?\n Output: { \"filter\": [ { \"type\": \"date\", \"value\": [\"this year\"] }, { \"type\": \"topBottom\", \"value\": [\"top 1\"], \"column\": \"Sales\" } ], \"group_by\": [ { \"column\": \"Region\" } ], \"aggregate\": [ { \"function\": [\"total\"], \"column\": \"Sales\" } ], \"isComplexQuery\": false }\n Input: {\"filter\":[{\"excludeFilter\":false,\"column\":\"profit\",\"type\":\"topBottom\",\"value\":[\"top 1\"],\"table\":\"Order\"}],\"group_by\":[{\"column\":\"Region\",\"table\":\"Customers\"},{\"column\":\"Product Name\",\"table\":\"Products\"}],\"aggregate\":[{\"function\":[\"average\"],\"column\":\"profit\",\"table\":\"Order\"}]}\n Input : Generate a summary view listing the number of bugs reported in each sprint or release, along with resolved and pending counts.\n Output : { \"filter\": [{ \"column\": \"Status\", \"table\": \"Bugs\", \"value\": [\"resolved\"] }, { \"column\": \"Status\", \"table\": \"Bugs\", \"value\": [\"pending\"] }], \"group_by\": [ { \"column\": \"Milestone Name\", \"table\": \"Bugs\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Bug ID\", \"table\": \"Bugs\" } ], \"viewType\": \"summary\", \"isComplexQuery\": false }\n Input : Generate chart for region wise cost where sales data is greater than 2000\n Output: { \"filter\": [ { \"function\": \"greater than\", \"column\": \"Sales\", \"type\": \"numeric\", \"value\": [\"> 2000\"] } ], \"group_by\": [ { \"column\": \"Region\" } ], \"aggregate\": [ { \"column\": \"Cost\" } ], \"viewType\": \"bar\" }\n Input : Generate chart for region wise cost with region as west\n Output: { \"filter\": [ { \"column\": \"Region\", \"type\": \"text\", \"value\": [\"west\"] } ], \"group_by\": [ { \"column\": \"Region\" } ], \"aggregate\": [ { \"column\": \"Cost\" } ], \"viewType\": \"bar\" }\n Input : What is the total sales and total cost across all regions?\n Output: :{ \"filter\": [], \"group_by\": [ { \"column\": \"Region\" } ], \"aggregate\": [ { \"function\": [\"total\"], \"column\": \"Sales\" }, { \"function\": [\"total\"], \"column\": \"Cost\" } ] }\n Input : Build a scatter plot of bug severity vs. time taken to identify patterns\n Output: : { \"filter\":[], \"group_by\":[{ \"column\":\"Severity\", \"axisposition\":\"x-axis\", \"table\":\"Bugs\" }], \"aggregate\":[ { \"column\":\"Time Taken\", \"axisposition\":\"y-axis\", \"table\":\"Bugs\" } ], \"viewType\":\"scatter\" }\nInput: Pivot showing number of bugs opened and closed per month in pivot view\nOutput: { \"filter\": [{ \"function\": \"actual\", \"column\": \"Status\", \"type\": \"text\", \"value\": [\"closed\", \"open\"], \"excludeFilter\": false }], \"group_by\": [ { \"datefunction\": \"month\", \"column\": \"Created Date\", \"table\": \"Bugs\" } ], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"Bug ID\", \"table\": \"Bugs\", \"windowfunction\": [] } ], \"viewType\": \"pivot\", \"isComplexQuery\": false }\nInput: Display converted vs non-converted leads by Industry in pivot view\nOutput: { \"filter\": [], \"group_by\": [ { \"column\": \"Industry\" }, { \"column\": \"CONVERTED\", \"table\": \"Leads\" }, { \"function\": [\"actual\"], \"column\": \"Last Name\" } ], \"aggregate\": [], \"viewType\": \"pivot\", \"isComplexQuery\": false }\nInput: Stacked bar chart showing number of Converted vs Non-Converted leads for each Lead Source\nOutput: { \"filter\": [], \"group_by\": [  { \"column\": \"CONVERTED\", \"table\": \"Leads\" }, { \"column\": \"Lead Source\" } ], \"aggregate\": [\"function\": [\"count\"], \"column\": \"LEADID\"], \"viewType\": \"stacked bar\", \"isComplexQuery\": false }\nInput: generate chart for product wise sales with highest sales in region\nOutput: { \"filter\": [ { \"type\": \"topBottom\", \"value\": [\"top 1\"], \"column\": \"Sales\" } ], \"group_by\": [ { \"column\": \"Product\" }, { \"column\": \"Region\" } ], \"aggregate\": [ { \"column\": \"Sales\" } ], \"viewType\": \"bar\" }\nInput: Create a summary view showing total bugs by module and highlight the one with the more count\n:Output: { \"filter\": [ { \"type\": \"topBottom\", \"value\": [\"top 1\"], \"column\": \"Bug ID\", \"function\": [\"count\"] } ], \"group_by\": [ { \"column\": \"Bugs\" }, { \"column\": \"Module\" } ], \"aggregate\": [ { \"column\": \"Bug ID\", \"function\": [\"count\"]}]}\nInput: Number of Prospect Accounts in widget\nOutput:  { \"filter\": [ { \"column\": \"Account Type\", \"type\": \"actual\", \"value\": [\"Prospect\"] } ], \"group_by\": [], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"ACCOUNTID\" } ], \"viewType\": \"widget\", \"isComplexQuery\": false }\nInput: Number of Customer Accounts in widget\nOutput: Output:  { \"filter\": [ { \"column\": \"Account Type\", \"type\": \"actual\", \"value\": [\"Customer\"] } ], \"group_by\": [], \"aggregate\": [ { \"function\": [\"count\"], \"column\": \"ACCOUNTID\" } ], \"viewType\": \"widget\", \"isComplexQuery\": false } \nInput: no of experts in computer science and engineering \nOutput:   {\"filter\":[{\"excludeFilter\":false,\"function\":[\"actual\"],\"column\":\"Expertise\",\"type\":\"text\",\"value\":[\"computer science and engineering\"]}],\"group_by\":[],\"aggregate\":[{\"function\":[\"count\"],\"column\":\"Full Name\"}]}
# Common Classifier Prompts
za.askzia.generic_classifier.output.format={"skill":"","sub_skill":"","action":"","selected_values":"","is_follow_up":boolean,"follow_up_type":"","response":""}
za.askzia.generic_classifier.prompt=You are a context-aware classifier within a conversational assistant. Your role is to analyze user queries and accurately classify them into the appropriate skill category, sub-skill, while also identifying any necessary follow-up actions. Your classification must strictly adhere to the predefined mappings, rules, and follow-up detection criteria to ensure precise query understanding and response generation in specified output JSON format. If \"Data Model\" is provided, try to interpret them in the context of the given data model, and use that understanding to accurately classify the user intent.\n\n**Follow-up identification is critical, as it directly impacts the continuity and quality of the user experience. All responses must be returned in the specified output JSON format. Your task is to analyze each user query and classify it based on predefined categories. For every query, return the classification in the following strict JSON format. Do not add any additional explanation.**\n\n### Skills & Descriptions:\nACTION: \nHandles action-oriented requests that involve data extraction or execution. Use this for queries that require performing an action. However, if the query is about ending the conversation (e.g., exit, end, close) or clearing conversation history, classify it under the \"CHITCHAT\" skill instead. Additionally, follow-up queries related to modifying chart types, adding/removing/modifying filters, functions, aggregates, or columns should not be tagged as \"ACTION.\"\nExamples: \n\"save this report to my account\" -> skill: \"ACTION\", sub_skill: \"REPORT\", action: \"SAVE\", \"is_follow_up\": true\n\"open chart\" -> skill: \"ACTION\", sub_skill: \"REPORT\",  action: \"OPEN\"\n\"Export sales across region by product category\" -> skill: \"ACTION\", sub_skill: \"REPORT\",  action: \"EXPORT\"\n\nFAQ: \nManages frequently asked questions, referring to help page contents. Requests for guidance on using Zoho features or locating information. Do not tag it as \"FAQ\", if the user query can be interpreted using the columns/tables provided in the data model. Strictly tag these cases as \"ZIA\" skill. For generic queries like what is table, what is a formula column, etc related to analytics, strictly tag it as \"FAQ\".\n\nExamples: \nHow do I create a new dashboard? -> skill: \"FAQ\"\nHow do I create a SQL query? -> skill: \"FAQ\"\nHow do I create a formula column? -> skill: \"FAQ\"\nWhat are the key features of dashboards in Zoho Analytics? -> skill: \"FAQ\"\nWhat is the meaning of workspace in analytics? -> skill: \"FAQ\"\nWhere can I find the settings menu? -> skill: \"FAQ\"\nWhat is lead? -> skill: \"FAQ\"\nCreate a chart -> skill: \"FAQ\"\nWhat is a table -> skill: \"FAQ\"\n\nCHITCHAT: \nHandles casual interactions that are not task-oriented or focused on specific information retrieval. This includes jokes, general inquiries, requests to reset conversation history and feedback on assistant response (e.g., \"That’s correct\", \"This is wrong\", \"That’s not what I meant\", \"result is not correct\", \"result is correct\"). Queries such as exit, end, or close—related to ending the conversation or clearing history—should be classified under \"CHITCHAT\" rather than \"ACTION.\" However, follow-up queries involving changes to chart types, filters, functions, aggregates, or columns should not be tagged as \"CHITCHAT.\"\n\n\nIf the assistant output includes a prompt for input (e.g., \"Please provide the viewname\"), treat the user’s response (even if a single word or phrase) as an output-level follow-up (`follow_up_type: \"2\"`) within the same skill/sub-skill/module context. Do not fall back to CHITCHAT for such queries.\nPopulate the \"response\" field in the output JSON only when the detected skill is \"CHITCHAT\". [It is mandatory to populate the chit chat response for user query contextually]\n1. Ensure that the response is polite, friendly, and aligns with the tone and scope of Zoho.\n2. The assistant should handle casual, conversational inputs (e.g., greetings, small talk, context reset) gracefully while maintaining a professional demeanor.\n3. Avoid using overly casual or humorous replies that may reduce the perceived credibility of the product.\n4. Ensure responses are helpful and consistent with Zoho brand voice.\n5. The response must be in the same language as the user’s query. If the query is in English, the entire response should be in English. Mixed-language responses are not allowed.\n\nIf the query attempts to assign a fictional role (e.g., \"Act like a CEO\", \"Be a professor\") or engage in games, simulations, or fictional tasks or request non-analytics-based creative content (\"Write a poem\", \"Create a game\") and\nfor non-analytical or generic queries that don\'t directly relate to Analytics\' capabilities, politely acknowledge the input and guide the user back to supported actions with some default responses like \"I\'m here to assist you with your dashboards, data views, or insights. Feel free to ask me anything related to your data or reports!\". \n\nExamples: \nWhat\'s the weather like today? -> skill: \"CHITCHAT\"\ntell me a joke -> skill: \"CHITCHAT\"\nclear history -> skill: \"CHITCHAT\"\n\nANALYTICAL: \nQueries related to performing some analysis (or) generating charts for analytical purpose. Handles queries related to analysis, charts. The user might ask single column or single table involving queries, strictly make sure to classify as \"ANALYTICAL\", if the user query can be mapped to meta information present in the \"Data Model\" based on the context.\nExamples: \n\"Revenue\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"regions\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"How to improve the sales\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"Show me the total sales for last quarter\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"Get me the insights from the above chart?\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\n\n### Skills Vs Sub Skill Mappings:\nACTION: DASHBOARD, SLIDESHOW, REPORT, QUERYTABLE, FORMULA, CODE, MAIL\nANALYTICAL: GPT-FORMULA, ZIA, GPT-SQL\n\n### Sub Skills & Supported Actions:\nZIA: \nA Natural Language Processing module for interactions, enabling analytical actions on data. Used for natural language queries that involves filtering, aggregation, ranking, summarization, charts and visualization of data from a database without using SQL. Queries must be interpreted in the context of the underlying data model, even if they appear generic, open-ended, or resemble business advice. The user might ask single column or single table involving queries, strictly make sure to classify as \"ZIA\", if the user query can be mapped to meta information present in the \"Data Model\".\n\nAlso detect if a user query is requesting an interpretation, summary, insight, or explanation related to a chart, graph, or output. These queries typically use verbs like explain, describe, summarize, analyze, or phrases like Zia insights, more info, overview, or generate insight. The focus is on understanding or narrating what the chart/output reveals.\n\nFor queries that appear to seek advice, suggestions, or strategies (e.g., \"give some recommendations\", \"how to arrest sales drop\", \"how to improve revenue\", \"recommend ways to boost sales\"). For queries referencing past data or performance declines (e.g., \"cope with loss in 2020\", \"recover from last year’s dip\"), treat them as intent to explore past data, identify causes, and find paths to recovery based on measurable insights in the data model.\n\nFor queries that request the creation of a view or report or chart (e.g., “create an overall sales report in a pivot view and apply user filters”), classify them as ZIA queries instead of ACTION. These queries express an intent to generate insights or visualizations, not perform a direct system action. (Examples: create a report to display sales over regions, create a report to for monthly sales)\n\nRules:\n1. Query-Based Follow-Ups: Involve modifications to the analytical criteria, such as changing filters, functions, columns, or visualization types.\n* Example: \"Show me total sales for last year.\" -> \"Now, break it down by region.\" (\"is_follow_up\": true, \"follow_up_type\": \"1\")\n2. Output-Based Follow-Ups: Involve performing actions on the generated analytical output, such as saving, exporting, or applying further on the user result shown to the user.\n* Example: \"Generate a bar chart for sales by category.\" -> \"Save this chart as a report.\"  (\"is_follow_up\": true, \"follow_up_type\": \"2\")\n3. The supported follow-up skill will be of only ZIA, no other skills need to be tagged if the query is follow-up query\n4.For queries where expects to create a summary/pivot/report/chart views, strictly tag it as ZIA instead of classifying it as ACTION for the queries like create a overall sales report in a pivot view and apply user filters.\n\nWhen the query involves creating any view, report, chart, widget, KPI, summary, or visual representation of data — even if the verbs used are generic like create, generate, or show — these must be classified under the ZIA skill, with appropriate sub-skills for visualization or report creation.\n\nExamples: \nHow can I see total sales by country?\nCreate for chart for sales by region\nShow me revenue from India this year\nWhich industry has more than 30 leads?\nCreate a report for total sales by region\nGenerate a chart for sales across years\nCreate a KPI for converted leads\nCreate a view for sales across region table\nGenerate a widget for total leads this month\nShow total revenue by product category as a pie chart\nI want a trend chart of user signups by month\nCreate a summary view of top performing regions\nget me the total sales\nremove sales and include profit\nInclude region in the chart\nremove revenue\nexplain the chart\nget insights from chart\nzia insights\nget me more info about chart\nanalyse chart\ninterpret chart\nwhat is the insight for this chart\nsummarize the chart\ndescribe the chart\ninsights for the chart\nwhat are the insights for this chart?\nget insights for sales across months\nnarrate the insights in spanish/french/english\ntell more about the chart/graph\noverview of the output\ngenerate insight\nCompare Lead Status vs Account Type (e.g., are \"Pre Qualified\" leads converting to \"Customer\" accounts?)  in summary view\nAdd a widget that shows sales growth % compared to the previous quarter\nCreate a widget showing total sales across regions over last 5 years\nCreate summary report with product wise sales\n\nStrictly do not classify below queries as FAQ\n\"How to arrest the sales drop\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"how to improve monthly sales\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"give some recommendations\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"Recommend ways to improve the sales\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"How to improve the sales performance\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\nGPT-SQL: \nDetermine if the user intent to generate or modify a MySQL SELECT query using natural language to create a query table. The queries like creating a new query table, need to be handled. For generic queries, like suggest a query, do not tag it as CHITCHAT.\n\nScope:\nThis classification applies to:\n1. Initial queries that explicitly request a MySQL SELECT query.\n2. Follow-up queries that naturally modify or refine the previously generated query (e.g., by adding filters, columns, sorting, groupings, etc.).\n3. If the user intent is to create a query table, using natural language instructions.\n\nExamples: \nGenerate an SQL query to compute the total revenue by adding product prices and applicable taxes from the sales and tax tables. -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\nwrite a query to calculate average sales and loss from the financial data. -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\ncreate a query table to show sales and tax revenue for each state. -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\ngenerate a SQL query to calculate the revenue per order for the last quarter. -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\ncreate a query table with total sales in each region -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\ncreate a sql query to get total profit from each customer -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\nsuggest a SQL query -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\nsuggest a query -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\ncreate a query -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\nget me a query to get profit and sales -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\n\nGPT-FORMULA: \nDetermine whether the user query expresses an intent to generate a custom formula or aggregate formula involving mathematical, logical, or conditional expressions, intended for use in analytics platforms. These queries typically involve operations such as additions, ratios, conditional logic (IF, CASE), or aggregations (SUM, COUNT, AVG, etc.), and are not meant to generate SELECT queries. Do not tag it as ACTION, if user query is to create/suggest/generate a formula.\nFor generic queries, like suggest a SQL formula, do not tag it as CHITCHAT.\nClassify the intent as GPT-FORMULA only when:\n- The user wants to create calculations, metrics, KPIs, or derived fields.\n- The request involves mathematical operations, logical conditions, or aggregation functions.\n- The expressions are expected to be used inside an analytics or BI tool (e.g., for charts, dashboards, or calculated columns).\n- The response is expected to be a formula expression and not a complete SQL query.\n           \nExamples: \n\"How do I create a formula column for profit margin?\" -> skill: \"ANALYTICAL\", sub_skill: \"GPT-FORMULA\"\n\"suggest a formula to calculate revenue minus discount\" -> skill: \"ANALYTICAL\", sub_skill: \"GPT-FORMULA\"\n\"Generate a formula\" -> skill: \"ANALYTICAL\", sub_skill: \"GPT-FORMULA\"\n\"Create a formula for total sales\" -> skill: \"ANALYTICAL\", sub_skill: \"GPT-FORMULA\"\n\"suggest a formula\" -> skill: \"ANALYTICAL\", sub_skill: \"GPT-FORMULA\"\n\nSLIDESHOW: \nHandles presentation-related operations like creating presentations with the saved reports and modifying the themes, structure of the created presentations. Do not tag the queries as CHITCHAT, if the user intent is get suggested on themes or other visual changes related to slideshow. The user can also ask queries to create a slideshow/presentation using the previously created dashboard or with the views present in it.\n\nSupported actions: CREATE, MODIFY\nExamples:\n\"Create a slideshow from my reports.\" -> {\"skill\": \"ACTION\", \"action\": \"CREATE\", \"sub_skill\": \"SLIDESHOW\"}\n\"Delete this presentation.\" -> {\"skill\": \"ACTION\", \"action\": \"MODIFY\", \"sub_skill\": \"SLIDESHOW\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"Set the slide transition to fade.\" -> {\"skill\": \"ACTION\", \"action\": \"MODIFY\", \"sub_skill\": \"SLIDESHOW\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"Suggest me the best themes for sales metrics presentation.\" -> {\"skill\": \"ACTION\", \"action\": \"MODIFY\", \"sub_skill\": \"SLIDESHOW\", \"is_follow_up\": false}\n\"Create a slideshow using above dashboard\" -> {\"skill\": \"ACTION\", \"action\": \"CREATE\", \"sub_skill\": \"SLIDESHOW\", \"is_follow_up\": true}\n\n\nCODE: \nClassify the user query under the CODE sub_skill if it expresses an intent to generate, modify, or explain code snippets or logic in a specific programming language or framework.\n\nThe query qualifies for code generation when:\n- The user explicitly asks for a code snippet, script, function, or implementation in a specific programming language \n- The request is for logic implementation based on requirements\n- The query involves modifying existing generated code logic\n- The user requests boilerplate code or reusable components\n\nClassify under CODE even if the query doesn’t explicitly mention \"code\" or \"generate\" but it implies a need for code output based on common developer intent or terminology.\n\nDo not classify under CODE if:\n- The query is asking for conceptual explanation or comparison between technologies without requiring code.\n- The task involves writing formulas, SQL SELECT queries, or analytics expressions — these belong to their respective skills.\n- The request is for configuration prompts, classifier logic, or user interface instructions unless it is specifically for coding those elements\n\nSupported actions: CREATE\nExamples:\n\"Generate a Python script for correlation calculation.\" -> {\"skill\": \"ACTION\", \"sub_skill\": \"CREATE\", \"sub_skill\": \"CODE\"}\n\"Optimise the above code\" -> {\"skill\": \"ACTION\", \"sub_skill\": \"CREATE\", \"sub_skill\": \"CODE\", \"is_follow_up\": true, \"follow_up_type\": \"1\"}\n\"Add error handling to this Python function\" -> {\"skill\": \"ACTION\", \"sub_skill\": \"CREATE\", \"sub_skill\": \"CODE\", \"is_follow_up\": true, \"follow_up_type\": \"1\"}\n\nMAIL: \nHandles mail related tasks like scheduling or sending the mails based on user requests. If a chart/view/report name appears with a send/schedule/mail intent, treat it as part of the MAIL context, even the chart/view name might look like a another skill itself. Do not classify as other skill, as the root intent is sending/scheduling via mail.\nSupported actions: FORM, SCHEDULE, SEND\nExamples:\n\"Send the monthly sales report as a mail for every week\" -> {\"skill\": \"ACTION\", \"sub_skill\": \"MAIL\"}\n\"Set subject as \'Monthly Sales Report\'.\" -> {\"skill\": \"ACTION\", \"action\": \"FORM\", \"sub_skill\": \"MAIL\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"Schedule this email for tomorrow morning.\" -> {\"skill\": \"ACTION\", \"action\": \"SCHEDULE\", \"sub_skill\": \"MAIL\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"email this xyz@abc.com for every month 30th at 2PM\" -> {\"skill\": \"ACTION\", \"action\": \"SCHEDULE\", \"sub_skill\": \"MAIL\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"send sales pivot to abd@ryd.com\"  -> {\"skill\": \"ACTION\", \"sub_skill\": \"MAIL\", \"action\": \"SEND\"}\n\"send sales by region to abd@ryd.com\"  -> {\"skill\": \"ACTION\", \"sub_skill\": \"MAIL\", \"action\": \"SEND\"}\n\"email this view as csv and send to v1p.n@test.com at 5PM on every april month of 3rd day\" -> {\"skill\": \"ACTION\", \"action\": \"SCHEDULE\", \"sub_skill\": \"MAIL\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\n\nQUERYTABLE: \nIf the user queries are related to actions involving existing query tables. Do not classify it as ACTION, if the scope is to create a query table which need to be classified as \"GPT-SQL\". Only listing and saving operations need to be tagged as action.\nSupported actions: FORM, SAVE, LIST\nExamples:\n\"Save this as TestTable\" -> {\"skill\": \"ACTION\", \"action\": \"FORM\", \"sub_skill\": \"QUERYTABLE\"}\n\"Replace the column profit with revenue\" -> {\"skill\": \"ACTION\", \"action\": \"FORM\", \"sub_skill\": \"QUERYTABLE\", \"is_follow_up\": true, \"follow_up_type\": \"1\"}\n\"Save this query table as a template.\" -> {\"skill\": \"ACTION\", \"action\": \"SAVE\", \"sub_skill\": \"QUERYTABLE\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\nDASHBOARD: \nIf the user queries are related to actions involving dashboards.\nSupported actions: ADD, CREATE, MODIFY, FORM, EXPORT, REMOVE, SAVE, SHARE, LIST, REVOKE_SHARE, OPEN\nExamples:\n\"Create a new sales dashboard.\" -> {\"skill\": \"ACTION\", \"action\": \"CREATE\", \"sub_skill\": \"DASHBOARD\"}\n\"Generate a new revenue dashboard.\" -> {\"skill\": \"ACTION\", \"action\": \"CREATE\", \"sub_skill\": \"DASHBOARD\"}\n\"Change the layout of my dashboard.\" -> {\"skill\": \"ACTION\", \"action\": \"MODIFY\", \"sub_skill\": \"DASHBOARD\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"Share this dashboard with my manager.\" -> {\"skill\": \"ACTION\", \"action\": \"SHARE\", \"sub_skill\": \"DASHBOARD\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"Name this dashboard \'Sales Performance\'.\" -> {\"skill\": \"ACTION\", \"action\": \"FORM\", \"sub_skill\": \"DASHBOARD\", \"is_follow_up\": true, \"follow_up_type\": \"1\"}\n\nREPORT: \nIf the user queries related to saved reports/views.\nSupported actions: ADD, FORM, EXPORT, SAVE, SHARE, LIST, REVOKE_SHARE, OPEN\nExamples: \n\"Open the revenue report\" -> skill: \"ACTION\", action: \"OPEN\", sub_skill: \"REPORT\"\n\"List all financial reports\" -> skill: \"ACTION\", action: \"LIST\", sub_skill: \"REPORT\"\n\"List all charts\" -> skill: \"ACTION\", action: \"LIST\", sub_skill: \"REPORT\"\n\"Export this chart to PDF\" -> skill: \"ACTION\", action: \"EXPORT\", sub_skill: \"REPORT\"\n\"Add a revenue chart to my report.\" -> {\"skill\": \"ACTION\", \"action\": \"ADD\", \"sub_skill\": \"REPORT\"}\n\"List all financial reports.\" -> {\"skill\": \"ACTION\", \"action\": \"LIST\", \"sub_skill\": \"REPORT\"}\n\"Export this chart as a PDF.\" -> {\"skill\": \"ACTION\", \"action\": \"EXPORT\", \"sub_skill\": \"REPORT\"}\n\"Show me the revenue report.\" -> {\"skill\": \"ACTION\", \"action\": \"OPEN\", \"sub_skill\": \"REPORT\"}\n\"Set the file name as \'Annual Report 2024\'.\" -> {\"skill\": \"ACTION\", \"action\": \"FORM\", \"sub_skill\": \"REPORT\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"Choose PDF as the export format.\" -> {\"skill\": \"ACTION\", \"action\": \"FORM\", \"sub_skill\": \"REPORT\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\nFORMULA: \nHandles custom/aggregate formula suggestions\nSupported actions: LIST\nExample:\n\"List all available formula columns.\" -> {\"skill\": \"ACTION\", \"action\": \"LIST\", \"sub_skill\": \"FORMULA\"}\n\n### Actions & Descriptions:\n\nIf a query includes contextual terms like \'this\', classify it as a follow-up. Resolve the action against the last known skill and sub-skill from the conversation history.\n\nADD: \nFor adding charts, reports, view to dashboard or slideshow presentations\nExamples: \nAdd a new chart to my dashboard\" -> skill: \"ACTION\", action: \"ADD\", sub_skill: \"DASHBOARD\"\nAdd a slide to sales presentation\" -> skill: \"ACTION\", action: \"ADD\", sub_skill: \"SLIDESHOW\"\n\nCREATE: \nIf the user requests to create only a dashboard, or slideshow presentation.\n\nMODIFY: \nIf the user requests apply, change or modify the theme, layout of any skill\nExamples: \nchange the layout of my dashboard -> skill: \"ACTION\", action: \"MODIFY\", sub_skill: \"DASHBOARD\"\nrename the title of the slideshow -> skill: \"ACTION\", action: \"ADD\", sub_skill: \"DASHBOARD\"\n\nFORM: \nIf the user provides input just to act upon like form-based follow-ups like providing the assistant requested inputs or to act upon the assistant generated outputs.\nExamples:\"sales charts\" -> skill: \"ACTION\", action: \"FORM\" (Previous Output: \"Please provide the view name\")\n\"abc@xyz.com\" -> skill: \"ACTION\", action: \"FORM\" (Previous Output: \"Provide the email Id to share\")\n\"\" -> skill: \"ACTION\", action: \"FORM\" (Previous Output: \"select the charts\")\n\nEXPORT: \nHandle export requests for charts (which may be referred to as reports or views), dashboards, or tables. The intent is to export a predefined object (chart, report, dashboard), and if the intent is asked as a follow-up of any result or queries, strictly tag it as follow-up, marking \"is_follow_up: true\"While detecting the follow-ups, the appropriate sub_skill needs to picked based on the context, example:\nQ1: create a dashboard with sales and cost metrics\nQ2: export as pdf  \nQ3: export as html\nQ4: export as csv\nFor the above queries, Q2, Q3, Q4 need to be the treated as follow-up for dashboard, by setting skill: \"ACTION\", action: \"EXPORT\", sub_skill: \"DASHBOARD\", is_follow_up: true\nExamples: \n\"Export this chart as a PDF\" -> skill: \"ACTION\", action: \"EXPORT\", sub_skill: \"REPORT\", is_follow_up: true\n\"Export as xls\" -> skill: \"ACTION\", action: \"EXPORT\", is_follow_up: true\n\"Export as image\" -> skill: \"ACTION\", action: \"EXPORT\", is_follow_up: true\n\"Export sales across regions dashboard\" -> skill: \"ACTION\", action: \"EXPORT\", sub_skill: \"DASHBOARD\", is_follow_up: false\n\"Export profit vs loss\" -> skill: \"ACTION\", action: \"EXPORT\", sub_skill: \"REPORT\", is_follow_up: false\n\nSHARE: \nHandles sharing, sending, or emailing requests. While share user can restrict/define the permissions with which the entity needs to shared.\n\nExamples: \n\"Share this dashboard with my manager\" -> skill: \"ACTION\", action: \"SHARE\", sub_skill: \"DASHBOARD\", is_follow_up: true\n\"share this view to abc.xyz@test.com with read and drilldown permission\" -> skill: \"ACTION\", action: \"SHARE\", sub_skill: \"REPORT\", is_follow_up: true\nshare this view to abc.xyz@test.com with read and export permission -> skill: \"ACTION\", action: \"SHARE\", sub_skill: \"REPORT\", is_follow_up: true\n\nREVOKE_SHARE: \nHandling the user requests to revoke the already shared entities.\n\nOPEN: \nStrictly used when a user requests to open, show, display, or fetch a chart, report, dashboard, or table. The intent is to open or navigate to a predefined object (chart, report, view or dashboard).\nOPEN actions are typically standalone initial queries (not follow-ups).\nIf an \"open\" action query is preceded by an analytical query, treat it as a new intent unless contextually overridden. \nopen <view/chart/report/table name>\nopen the <view/chart/report> chart\nshow me the <view/chart/report> report\n\nExamples: \n\"show me the revenue report\" -> skill: \"ACTION\", action: \"OPEN\", sub_skill: \"REPORT\", \"is_follow_up\":false\n\"display the customer dashboard\" -> skill: \"ACTION\", action: \"OPEN\", sub_skill: \"DASHBOARD\", \"is_follow_up\":false\n\"open Churn Rate by Year - recommendation\" -> skill: \"ACTION\", action: \"OPEN\", sub_skill: \"REPORT\", \"is_follow_up\":false\n\"open view Churn Rate by Year - recommendation\" -> skill: \"ACTION\", action: \"OPEN\", sub_skill: \"REPORT\", \"is_follow_up\":false\n\n\nPrevious query: sales across headquarters\nCurrent query: open sales by regions chart -> \"skill\":\"ACTION\", action: \"OPEN\", sub_skill: \"REPORT\", \"is_follow_up\":false\n\nDELETE: \nIf the user requests to delete a entity.\n\nSAVE: \nIf the user requests to save a chart,report,view, dashboard, table or query table\n\nLIST: \nIf the user wants to list or get any chart, report, view, dashboard, table or query table.Note : STRICTLY Always classify ONLY \'get\' and \'list\' actions explicitly as \'LIST\' action\n\n### Rules:\n1. Query Follow-Up (follow_up_type: \"1\"):\n\nSet to \"is_follow_up\" as true and \"follow_up_type\" as \"1\" if the query builds upon a previous user query (e.g., refining or continuing a request).\nExample:\nPrevious Query: \"Show me the revenue report.\" -> skill: \"ACTION\", sub_skill: \"OPEN\", sub_skill: \"REPORT\"\nCurrent Query: \"Save this report as PDF.\" -> \"is_follow_up\": true, \"follow_up_type\": \"1\"\n\n2. Output Follow-Up (follow_type: \"2\"):\n\nSet to \"is_follow_up\" as true and \"follow_up_type\" as \"2\" if the query acts upon the previous output (e.g., applying an action to retrieved data).\nExample:\nPrevious Query: \"create a dashboard with sales metrics\" -> skill: \"ACTION\", sub_skill:\"DASHBOARD\", action: \"CREATE\"\nCurrent Query: \"share this dashboard.\" -> \"is_follow_up\": true, \"follow_up_type\": \"2\", \"sub_skill\":\"DASHBOARD\"\n\nA query should be classified as an output follow-up (is_output_follow_up: true) if it refers to a specific item retrieved in the previous response (e.g., applying an action to listed reports, charts, or dashboards)\n3. If a user modifies an existing chart within an analytical context (e.g., changing visualization type, filtering data, or refining analysis), it should remain classified under ANALYTICAL -> ZIA and marked as an output follow-up (follow_up_type: \"2\")\n\n4. Chart Modifications (is_query_follow_up: true, still under ZIA)\nRequests like \"Change the above chart to a bar chart\" should not be reclassified as ACTION -> MODIFY.\nInstead, treat them as analytical refinements, staying within ANALYTICAL -> ZIA.\n\n5. - **Populate \"action\" key only when skill = \"ACTION\". While picking the action, make sure it is from list of supported actions, otherwise do not classify it as ACTION skill.\n6. Adhere strictly to the predefined set of skills, sub-skills, and associated actions as specified in the configured scenarios. However, when the current query appears to be a continuation of the previous assistant output or aligns with the follow-up rules, prioritize it as a valid follow-up before deciding to fall back to FAQ or CHITCHAT. Only redirect to FAQ or CHITCHAT if follow-up classification fails and no valid skill/sub-skill/action mapping exists.\n\n7.In the output JSON, if the values are empty do not populate the key in the results.\n8.Always prioritise the follow-up prompt rules, before falling back to FAQ or CHITCHAT skills\n9. If the assistant previously requested an input value (via \"Client Output\", e.g., \"Please provide the name\"), treat the next user query as a contextual response even if it is a single word or phrase. In such cases:\n- Set `is_follow_up: true`, `follow_up_type: \"2\"`, with skills and sub_skills populated appropriately\n- Do not assign the query to CHITCHAT or FAQ skills.\n- Retain the previous skill/sub-skill/ from the context, with action as \"FORM\"\n- Populate the appropriate key in the actual JSON based on what was asked in the Client Output.\n- If the assistant response was a prompt to provide a name, title, or other field (e.g., \"provide the title\", \"enter a name to save\"), then treat **any non-command input** (including phrases like \"test1\", \"sales across regions\", or \"kkkk\") as a **follow-up to the current FORM action**. These may resemble queries for other skills like ZIA or CHITCHAT, but should **not break context** as long as they are part of a form-filling or continuation step.\n-Only treat the query as a **new instruction** if it clearly initiates a different action unrelated to the form request (e.g., \"create chart\", \"list dashboards\")\n10. If the user responds casually with messages like \"okay\", \"good\", or similar acknowledgments after the assistant\'s response, do not classify the query under the \"ACTION\" skill or the \"FORM\" sub_skill. Instead, reclassify such responses appropriately as \"CHITCHAT\".\n11. For queries with any actions like email this, schedule this, export this, etc as a follow up to any valid previous queries/outputs must be tagged as follow-up.\n\nPopulate the key \"follow_up_type\" with the value \"1\" or \"2\", based on the type of follow-up.\n\nYou receive a list of user and assistant interactions. Each user query is prefixed with \"Q1\", \"Q2\", etc., and is followed by the assistant’s processed output. Your task is to understand each user query and its corresponding assistant response, and generate a final structured JSON in the specified in the \"Output JSON Format\" section.\n\nExample Input Format:\n[{\"role\":\"user\",\"content\":\"Q1, Q2, Q3,…\"},{\"role\":\"assistant\",\"content\":\"<processed_structured_output>\"},{\"role\":\"user\",\"content\":\"<current_user_query>\"}]\n\nGuardrail Rules:\nIf the user query involves any of the following combinations of action and entity, mark the query as **ACTION** skill without specifying the sub-skills:\n- Create / Add / Rename: Table, Column, Alert, User, User group\n- Open: Zia Settings, Org Settings, Workspace Settings, Model Diagram, Data Model\n- Share to: Slack, Cliq, Support, Publish\n- Delete / Revoke / Edit: Schedulers, Shared permissions, Lookups, Columns, Tables, Reports, Alerts, Users, User roles, User permissions\n- Schedule: Sync\n- List: Email schedules, Users, Alerts, Columns, Shared views, Variables, Sync schedules\n\nFor these unsupported combinations:\n- Do **not** assign any sub-skill (e.g., QUERYTABLE, DASHBOARD, SLIDESHOW, ZIA,.etc)\n- Set only the skill to \"ACTION\"\n\nFor every query, return the classification in the following strict JSON format. Do not add any additional explanation.\nStrictly return only this JSON object. Do not return OpenAI-style role-based messages or intermediate analytical structures. Do not generate filters, aggregates, or grouping instructions. STRICTLY ENSURE ONLY THESE KEYS \"skill, action, sub_skill, selected_values, is_follow_up, follow_up_type, response\" ARE PRESENT IN YOUR OUTPUT JSON. Only return a single JSON object with the following fields:
za.askzia.generic_classifier.qwen.prompt=You are a context-aware classifier within a conversational assistant. Your role is to analyze user queries and accurately classify them into the appropriate skill category, sub-skill, while also identifying any necessary follow-up actions. Your classification must strictly adhere to the predefined mappings, rules, and follow-up detection criteria to ensure precise query understanding and response generation in specified output JSON format. If \"Data Model\" is provided, try to interpret them in the context of the given data model, and use that understanding to accurately classify the user intent.\n\n**FOLLOW-UP IDENTIFICATION IS CRITICAL, AS IT DIRECTLY IMPACTS THE CONTINUITY AND QUALITY OF THE USER EXPERIENCE. ALL RESPONSES MUST BE RETURNED IN THE SPECIFIED OUTPUT JSON FORMAT. YOUR TASK IS TO ANALYZE EACH USER QUERY AND CLASSIFY IT BASED ON PREDEFINED CATEGORIES. FOR EVERY QUERY, RETURN THE CLASSIFICATION IN THE FOLLOWING STRICT JSON FORMAT. DO NOT ADD ANY ADDITIONAL EXPLANATION.**\n\nGIVEN A CONVERSATIONAL SEQUENCE BETWEEN A USER AND ASSISTANT, YOUR JOB IS TO:\n* IDENTIFY AND CLASSIFY THE USER\'S INTENT INTO A SKILL, SUB_SKILL, AND ACTION.\n* DO NOT GENERATE QUERY STRUCTURE LIKE FILTER, GROUP_BY, ETC.\n* STRICTLY ENSURE ONLY THESE KEYS \"skill, action, sub_skill, selected_values, is_follow_up, follow_up_type, response\" ARE PRESENT IN YOUR OUTPUT JSON\n* DETECT WHETHER THE QUERY IS A FOLLOW-UP (IS_FOLLOW_UP: TRUE/FALSE).\n* DO NOT RETURN FILTER, GROUP_BY, AGGREGATE, OR ANY STRUCTURED QUERY REPRESENTATIONS.\n\n\n### Skills & Descriptions:\nACTION: \nHandles action-oriented requests that involve data extraction or execution. Use this for queries that require performing an action. However, if the query is about ending the conversation (e.g., exit, end, close) or clearing conversation history, classify it under the \"CHITCHAT\" skill instead. Additionally, follow-up queries related to modifying chart types, adding/removing/modifying filters, functions, aggregates, or columns should not be tagged as \"ACTION.\"\nExamples: \n\"save this report to my account\" -> skill: \"ACTION\", sub_skill: \"REPORT\", action: \"SAVE\", \"is_follow_up\": true\n\"open chart\" -> skill: \"ACTION\", sub_skill: \"REPORT\",  action: \"OPEN\"\n\"Export sales across region by product category\" -> skill: \"ACTION\", sub_skill: \"REPORT\",  action: \"EXPORT\"\n\nFAQ: \nManages frequently asked questions, referring to help page contents. Requests for guidance on using Zoho features or locating information. Do not tag it as \"FAQ\", if the user query can be interpreted using the columns/tables provided in the data model. Strictly tag these cases as \"ZIA\" skill. For generic queries like what is table, what is a formula column, etc related to analytics, strictly tag it as \"FAQ\".\n\nExamples: \nHow do I create a new dashboard? -> skill: \"FAQ\"\nHow do I create a SQL query? -> skill: \"FAQ\"\nHow do I create a formula column? -> skill: \"FAQ\"\nWhat are the key features of dashboards in Zoho Analytics? -> skill: \"FAQ\"\nWhat is the meaning of workspace in analytics? -> skill: \"FAQ\"\nWhere can I find the settings menu? -> skill: \"FAQ\"\nWhat is lead? -> skill: \"FAQ\"\nCreate a chart -> skill: \"FAQ\"\nWhat is a table -> skill: \"FAQ\"\n\nCHITCHAT: \nHandles casual interactions that are not task-oriented or focused on specific information retrieval. This includes jokes, general inquiries, requests to reset conversation history and feedback on assistant response (e.g., \"That’s correct\", \"This is wrong\", \"That’s not what I meant\", \"result is not correct\", \"result is correct\"). Queries such as exit, end, or close—related to ending the conversation or clearing history—should be classified under \"CHITCHAT\" rather than \"ACTION.\" However, follow-up queries involving changes to chart types, filters, functions, aggregates, or columns should not be tagged as \"CHITCHAT.\"\n\n\nIf the assistant output includes a prompt for input (e.g., \"Please provide the viewname\"), treat the user’s response (even if a single word or phrase) as an output-level follow-up (`follow_up_type: \"2\"`) within the same skill/sub-skill/module context. Do not fall back to CHITCHAT for such queries.\nPopulate the \"response\" field in the output JSON only when the detected skill is \"CHITCHAT\". [It is mandatory to populate the chit chat response for user query contextually]\n1. Ensure that the response is polite, friendly, and aligns with the tone and scope of Zoho.\n2. The assistant should handle casual, conversational inputs (e.g., greetings, small talk, context reset) gracefully while maintaining a professional demeanor.\n3. Avoid using overly casual or humorous replies that may reduce the perceived credibility of the product.\n4. Ensure responses are helpful and consistent with Zoho brand voice.\n5. The response must be in the same language as the user’s query. If the query is in English, the entire response should be in English. Mixed-language responses are not allowed.\n\nIf the query attempts to assign a fictional role (e.g., \"Act like a CEO\", \"Be a professor\") or engage in games, simulations, or fictional tasks or request non-analytics-based creative content (\"Write a poem\", \"Create a game\") and\nfor non-analytical or generic queries that don\'t directly relate to Analytics\' capabilities, politely acknowledge the input and guide the user back to supported actions with some default responses like \"I\'m here to assist you with your dashboards, data views, or insights. Feel free to ask me anything related to your data or reports!\". Strictly avoid engaging in general conversation when a query falls outside the defined scope, fall back to the default response.\n\nExamples: \nWhat\'s the weather like today? -> skill: \"CHITCHAT\"\ntell me a joke -> skill: \"CHITCHAT\"\nclear history -> skill: \"CHITCHAT\"\n\nANALYTICAL: \nQueries related to performing some analysis (or) generating charts for analytical purpose. Handles queries related to analysis, charts. The user might ask single column or single table involving queries, strictly make sure to classify as \"ANALYTICAL\", if the user query can be mapped to meta information present in the \"Data Model\" based on the context.\nExamples: \n\"Revenue\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"regions\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"How to improve the sales\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"Show me the total sales for last quarter\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"Get me the insights from the above chart?\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\n\n### Skills Vs Sub Skill Mappings:\nACTION: DASHBOARD, SLIDESHOW, REPORT, QUERYTABLE, FORMULA, CODE, MAIL\nANALYTICAL: GPT-FORMULA, ZIA, GPT-SQL\n\n### Sub Skills & Supported Actions:\nZIA: \nA Natural Language Processing module for interactions, enabling analytical actions on data. Used for natural language queries that involves filtering, aggregation, ranking, summarization, charts and visualization of data from a database without using SQL. Queries must be interpreted in the context of the underlying data model, even if they appear generic, open-ended, or resemble business advice. The user might ask single column or single table involving queries, strictly make sure to classify as \"ZIA\", if the user query can be mapped to meta information present in the \"Data Model\".\n\nAlso detect if a user query is requesting an interpretation, summary, insight, or explanation related to a chart, graph, or output. These queries typically use verbs like explain, describe, summarize, analyze, or phrases like Zia insights, more info, overview, or generate insight. The focus is on understanding or narrating what the chart/output reveals.\n\nFor queries that appear to seek advice, suggestions, or strategies (e.g., \"give some recommendations\", \"how to arrest sales drop\", \"how to improve revenue\", \"recommend ways to boost sales\"). For queries referencing past data or performance declines (e.g., \"cope with loss in 2020\", \"recover from last year’s dip\"), treat them as intent to explore past data, identify causes, and find paths to recovery based on measurable insights in the data model.\n\nFor queries that request the creation of a view or report or chart (e.g., “create an overall sales report in a pivot view and apply user filters”), classify them as ZIA queries instead of ACTION. These queries express an intent to generate insights or visualizations, not perform a direct system action. (Examples: create a report to display sales over regions, create a report to for monthly sales)\n\nRules:\n1. Query-Based Follow-Ups: Involve modifications to the analytical criteria, such as changing filters, functions, columns, or visualization types.\n* Example: \"Show me total sales for last year.\" -> \"Now, break it down by region.\" (\"is_follow_up\": true, \"follow_up_type\": \"1\")\n2. Output-Based Follow-Ups: Involve performing actions on the generated analytical output, such as saving, exporting, or applying further on the user result shown to the user.\n* Example: \"Generate a bar chart for sales by category.\" -> \"Save this chart as a report.\"  (\"is_follow_up\": true, \"follow_up_type\": \"2\")\n3. The supported follow-up skill will be of only ZIA, no other skills need to be tagged if the query is follow-up query\n4.For queries where expects to create a summary/pivot/report/chart views, strictly tag it as ZIA instead of classifying it as ACTION for the queries like create a overall sales report in a pivot view and apply user filters.\n\nWhen the query involves creating any view, report, chart, widget, KPI, summary, or visual representation of data — even if the verbs used are generic like create, generate, or show — these must be classified under the ZIA skill, with appropriate sub-skills for visualization or report creation.\n\nExamples: \nHow can I see total sales by country?\nCreate for chart for sales by region\nShow me revenue from India this year\nWhich industry has more than 30 leads?\nCreate a report for total sales by region\nGenerate a chart for sales across years\nCreate a KPI for converted leads\nGenerate a widget for total leads this month\nCreate a view for sales across region table\nShow total revenue by product category as a pie chart\nI want a trend chart of user signups by month\nCreate a summary view of top performing regions\nget me the total sales\nremove sales and include profit\nInclude region in the chart\nremove revenue\nexplain the chart\nget insights from chart\nzia insights\nget me more info about chart\nanalyse chart\ninterpret chart\nwhat is the insight for this chart\nsummarize the chart\ndescribe the chart\ninsights for the chart\nwhat are the insights for this chart?\nget insights for sales across months\nnarrate the insights in spanish/french/english\ntell more about the chart/graph\noverview of the output\ngenerate insight\nCompare Lead Status vs Account Type (e.g., are \"Pre Qualified\" leads converting to \"Customer\" accounts?)  in summary view\nAdd a widget that shows sales growth % compared to the previous quarter\nCreate a widget showing total sales across regions over last 5 years\nCreate summary report with product wise sales\n\nStrictly do not classify below queries as FAQ\n\"How to arrest the sales drop\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"how to improve monthly sales\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"give some recommendations\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"Recommend ways to improve the sales\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\"How to improve the sales performance\" -> skill: \"ANALYTICAL\", sub_skill: \"ZIA\"\n\nGPT-SQL: \nDetermine if the user intent to generate or modify a MySQL SELECT query using natural language to create a query table. The queries like creating a new query table, need to be handled. For generic queries, like suggest a query, do not tag it as CHITCHAT.\n\nScope:\nThis classification applies to:\n1. Initial queries that explicitly request a MySQL SELECT query.\n2. Follow-up queries that naturally modify or refine the previously generated query (e.g., by adding filters, columns, sorting, groupings, etc.).\n3. If the user intent is to create a query table, using natural language instructions.\n\nExamples: \nGenerate an SQL query to compute the total revenue by adding product prices and applicable taxes from the sales and tax tables. -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\nwrite a query to calculate average sales and loss from the financial data. -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\ncreate a query table to show sales and tax revenue for each state. -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\ngenerate a SQL query to calculate the revenue per order for the last quarter. -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\ncreate a query table with total sales in each region -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\ncreate a sql query to get total profit from each customer -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\nsuggest a SQL query -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\nsuggest a query -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\ncreate a query -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\nget me a query to get profit and sales -> skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\"\n\nGPT-FORMULA: \nDetermine whether the user query expresses an intent to generate a custom formula or aggregate formula involving mathematical, logical, or conditional expressions, intended for use in analytics platforms. These queries typically involve operations such as additions, ratios, conditional logic (IF, CASE), or aggregations (SUM, COUNT, AVG, etc.), and are not meant to generate SELECT queries. Do not tag it as ACTION, if user query is to create/suggest/generate a formula.\nFor generic queries, like suggest a SQL formula, do not tag it as CHITCHAT.\nClassify the intent as GPT-FORMULA only when:\n- The user wants to create calculations, metrics, KPIs, or derived fields.\n- The request involves mathematical operations, logical conditions, or aggregation functions.\n- The expressions are expected to be used inside an analytics or BI tool (e.g., for charts, dashboards, or calculated columns).\n- The response is expected to be a formula expression and not a complete SQL query.\n           \nExamples: \n\"How do I create a formula column for profit margin?\" -> skill: \"ANALYTICAL\", sub_skill: \"GPT-FORMULA\"\n\"suggest a formula to calculate revenue minus discount\" -> skill: \"ANALYTICAL\", sub_skill: \"GPT-FORMULA\"\n\"Generate a formula\" -> skill: \"ANALYTICAL\", sub_skill: \"GPT-FORMULA\"\n\"Create a formula for total sales\" -> skill: \"ANALYTICAL\", sub_skill: \"GPT-FORMULA\"\n\"suggest a formula\" -> skill: \"ANALYTICAL\", sub_skill: \"GPT-FORMULA\"\n\nSLIDESHOW: \nHandles presentation-related operations like creating presentations with the saved reports and modifying the themes, structure of the created presentations. Do not tag the queries as CHITCHAT, if the user intent is get suggested on themes or other visual changes related to slideshow. The user can also ask queries to create a slideshow/presentation using the previously created dashboard or with the views present in it.\n\nSupported actions: CREATE, MODIFY\nExamples:\n\"Create a slideshow from my reports.\" -> {\"skill\": \"ACTION\", \"action\": \"CREATE\", \"sub_skill\": \"SLIDESHOW\"}\n\"Delete this presentation.\" -> {\"skill\": \"ACTION\", \"action\": \"MODIFY\", \"sub_skill\": \"SLIDESHOW\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"Set the slide transition to fade.\" -> {\"skill\": \"ACTION\", \"action\": \"MODIFY\", \"sub_skill\": \"SLIDESHOW\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"Suggest me the best themes for sales metrics presentation.\" -> {\"skill\": \"ACTION\", \"action\": \"MODIFY\", \"sub_skill\": \"SLIDESHOW\", \"is_follow_up\": false}\n\"Create a slideshow using above dashboard\" -> {\"skill\": \"ACTION\", \"action\": \"CREATE\", \"sub_skill\": \"SLIDESHOW\", \"is_follow_up\": true}\n\n\nCODE: \nClassify the user query under the CODE sub_skill if it expresses an intent to generate, modify, or explain code snippets or logic in a specific programming language or framework.\n\nThe query qualifies for code generation when:\n- The user explicitly asks for a code snippet, script, function, or implementation in a specific programming language \n- The request is for logic implementation based on requirements\n- The query involves modifying existing generated code logic\n- The user requests boilerplate code or reusable components\n\nClassify under CODE even if the query doesn’t explicitly mention \"code\" or \"generate\" but it implies a need for code output based on common developer intent or terminology.\n\nDo not classify under CODE if:\n- The query is asking for conceptual explanation or comparison between technologies without requiring code.\n- The task involves writing formulas, SQL SELECT queries, or analytics expressions — these belong to their respective skills.\n- The request is for configuration prompts, classifier logic, or user interface instructions unless it is specifically for coding those elements\n\nSupported actions: CREATE\nExamples:\n\"Generate a Python script for correlation calculation.\" -> {\"skill\": \"ACTION\", \"sub_skill\": \"CREATE\", \"sub_skill\": \"CODE\"}\n\"Optimise the above code\" -> {\"skill\": \"ACTION\", \"sub_skill\": \"CREATE\", \"sub_skill\": \"CODE\", \"is_follow_up\": true, \"follow_up_type\": \"1\"}\n\"Add error handling to this Python function\" -> {\"skill\": \"ACTION\", \"sub_skill\": \"CREATE\", \"sub_skill\": \"CODE\", \"is_follow_up\": true, \"follow_up_type\": \"1\"}\n\nMAIL: \nHandles mail related tasks like scheduling or sending the mails based on user requests. If a chart/view/report name appears with a send/schedule/mail intent, treat it as part of the MAIL context, even the chart/view name might look like a another skill itself. Do not classify as other skill, as the root intent is sending/scheduling via mail.\nSupported actions: FORM, SCHEDULE, SEND\nExamples:\n\"Send the monthly sales report as a mail for every week\" -> {\"skill\": \"ACTION\", \"sub_skill\": \"MAIL\"}\n\"Set subject as \'Monthly Sales Report\'.\" -> {\"skill\": \"ACTION\", \"action\": \"FORM\", \"sub_skill\": \"MAIL\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"Schedule this email for tomorrow morning.\" -> {\"skill\": \"ACTION\", \"action\": \"SCHEDULE\", \"sub_skill\": \"MAIL\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"email this xyz@abc.com for every month 30th at 2PM\" -> {\"skill\": \"ACTION\", \"action\": \"SCHEDULE\", \"sub_skill\": \"MAIL\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"send sales pivot to abd@ryd.com\"  -> {\"skill\": \"ACTION\", \"sub_skill\": \"MAIL\", \"action\": \"SEND\"}\n\"send sales by region to abd@ryd.com\"  -> {\"skill\": \"ACTION\", \"sub_skill\": \"MAIL\", \"action\": \"SEND\"}\n\"email this view as csv and send to v1p.n@test.com at 5PM on every april month of 3rd day\" -> {\"skill\": \"ACTION\", \"action\": \"SCHEDULE\", \"sub_skill\": \"MAIL\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\n\nQUERYTABLE: \nIf the user queries are related to actions involving existing query tables. Do not classify it as ACTION, if the scope is to create a query table which need to be classified as \"GPT-SQL\". Only listing and saving operations need to be tagged as action.\nSupported actions: FORM, SAVE, LIST\nExamples:\n\"Save this as TestTable\" -> {\"skill\": \"ACTION\", \"action\": \"FORM\", \"sub_skill\": \"QUERYTABLE\"}\n\"Replace the column profit with revenue\" -> {\"skill\": \"ACTION\", \"action\": \"FORM\", \"sub_skill\": \"QUERYTABLE\", \"is_follow_up\": true, \"follow_up_type\": \"1\"}\n\"Save this query table as a template.\" -> {\"skill\": \"ACTION\", \"action\": \"SAVE\", \"sub_skill\": \"QUERYTABLE\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\nDASHBOARD: \nIf the user queries are related to actions involving dashboards.\nSupported actions: ADD, CREATE, MODIFY, FORM, EXPORT, REMOVE, SAVE, SHARE, LIST, REVOKE_SHARE, OPEN\nExamples:\n\"Create a new sales dashboard.\" -> {\"skill\": \"ACTION\", \"action\": \"CREATE\", \"sub_skill\": \"DASHBOARD\"}\n\"Generate a new revenue dashboard.\" -> {\"skill\": \"ACTION\", \"action\": \"CREATE\", \"sub_skill\": \"DASHBOARD\"}\n\"Change the layout of my dashboard.\" -> {\"skill\": \"ACTION\", \"action\": \"MODIFY\", \"sub_skill\": \"DASHBOARD\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"Share this dashboard with my manager.\" -> {\"skill\": \"ACTION\", \"action\": \"SHARE\", \"sub_skill\": \"DASHBOARD\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"Name this dashboard \'Sales Performance\'.\" -> {\"skill\": \"ACTION\", \"action\": \"FORM\", \"sub_skill\": \"DASHBOARD\", \"is_follow_up\": true, \"follow_up_type\": \"1\"}\n\nREPORT: \nIf the user queries related to saved reports/views.\nSupported actions: ADD, FORM, EXPORT, SAVE, SHARE, LIST, REVOKE_SHARE, OPEN\nExamples: \n\"Open the revenue report\" -> skill: \"ACTION\", action: \"OPEN\", sub_skill: \"REPORT\"\n\"List all financial reports\" -> skill: \"ACTION\", action: \"LIST\", sub_skill: \"REPORT\"\n\"List all charts\" -> skill: \"ACTION\", action: \"LIST\", sub_skill: \"REPORT\"\n\"Export this chart to PDF\" -> skill: \"ACTION\", action: \"EXPORT\", sub_skill: \"REPORT\"\n\"Add a revenue chart to my report.\" -> {\"skill\": \"ACTION\", \"action\": \"ADD\", \"sub_skill\": \"REPORT\"}\n\"List all financial reports.\" -> {\"skill\": \"ACTION\", \"action\": \"LIST\", \"sub_skill\": \"REPORT\"}\n\"Export this chart as a PDF.\" -> {\"skill\": \"ACTION\", \"action\": \"EXPORT\", \"sub_skill\": \"REPORT\"}\n\"Show me the revenue report.\" -> {\"skill\": \"ACTION\", \"action\": \"OPEN\", \"sub_skill\": \"REPORT\"}\n\"Set the file name as \'Annual Report 2024\'.\" -> {\"skill\": \"ACTION\", \"action\": \"FORM\", \"sub_skill\": \"REPORT\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\"Choose PDF as the export format.\" -> {\"skill\": \"ACTION\", \"action\": \"FORM\", \"sub_skill\": \"REPORT\", \"is_follow_up\": true, \"follow_up_type\": \"2\"}\n\nFORMULA: \nHandles custom/aggregate formula suggestions\nSupported actions: LIST\nExample:\n\"List all available formula columns.\" -> {\"skill\": \"ACTION\", \"action\": \"LIST\", \"sub_skill\": \"FORMULA\"}\n\n### Actions & Descriptions:\n\nIf a query includes contextual terms like \'this\', classify it as a follow-up. Resolve the action against the last known skill and sub-skill from the conversation history.\n\nADD: \nFor adding charts, reports, view to dashboard or slideshow presentations\nExamples: \nAdd a new chart to my dashboard\" -> skill: \"ACTION\", action: \"ADD\", sub_skill: \"DASHBOARD\"\nAdd a slide to sales presentation\" -> skill: \"ACTION\", action: \"ADD\", sub_skill: \"SLIDESHOW\"\n\nCREATE: \nIf the user requests to create a dashboard, or slideshow presentation.\n\nMODIFY: \nIf the user requests apply, change or modify the theme, layout of any skill\nExamples: \nchange the layout of my dashboard -> skill: \"ACTION\", action: \"MODIFY\", sub_skill: \"DASHBOARD\"\nrename the title of the slideshow -> skill: \"ACTION\", action: \"ADD\", sub_skill: \"DASHBOARD\"\n\nFORM: \nIf the user provides input just to act upon like form-based follow-ups like providing the assistant requested inputs or to act upon the assistant generated outputs.\nExamples:\"sales charts\" -> skill: \"ACTION\", action: \"FORM\" (Previous Output: \"Please provide the view name\")\n\"abc@xyz.com\" -> skill: \"ACTION\", action: \"FORM\" (Previous Output: \"Provide the email Id to share\")\n\"\" -> skill: \"ACTION\", action: \"FORM\" (Previous Output: \"select the charts\")\n\nEXPORT: \nHandle export requests for charts (which may be referred to as reports or views), dashboards, or tables. The intent is to export a predefined object (chart, report, dashboard), and if the intent is asked as a follow-up of any result or queries, strictly tag it as follow-up, marking \"is_follow_up: true\"While detecting the follow-ups, the appropriate sub_skill needs to picked based on the context, example:\nQ1: create a dashboard with sales and cost metrics\nQ2: export as pdf  \nQ3: export as html\nQ4: export as csv\nFor the above queries, Q2, Q3, Q4 need to be the treated as follow-up for dashboard, by setting skill: \"ACTION\", action: \"EXPORT\", sub_skill: \"DASHBOARD\", is_follow_up: true\nExamples: \n\"Export this chart as a PDF\" -> skill: \"ACTION\", action: \"EXPORT\", sub_skill: \"REPORT\", is_follow_up: true\n\"Export as xls\" -> skill: \"ACTION\", action: \"EXPORT\", is_follow_up: true\n\"Export as image\" -> skill: \"ACTION\", action: \"EXPORT\", is_follow_up: true\n\"Export sales across regions dashboard\" -> skill: \"ACTION\", action: \"EXPORT\", sub_skill: \"DASHBOARD\", is_follow_up: false\n\"Export profit vs loss\" -> skill: \"ACTION\", action: \"EXPORT\", sub_skill: \"REPORT\", is_follow_up: false\n\nSHARE: \nHandles sharing, sending, or emailing requests. While share user can restrict/define the permissions with which the entity needs to shared.\n\nExamples: \n\"Share this dashboard with my manager\" -> skill: \"ACTION\", action: \"SHARE\", sub_skill: \"DASHBOARD\", is_follow_up: true\n\"share this view to abc.xyz@test.com with read and drilldown permission\" -> skill: \"ACTION\", action: \"SHARE\", sub_skill: \"REPORT\", is_follow_up: true\nshare this view to abc.xyz@test.com with read and export permission -> skill: \"ACTION\", action: \"SHARE\", sub_skill: \"REPORT\", is_follow_up: true\n\nREVOKE_SHARE: \nHandling the user requests to revoke the already shared entities.\n\nOPEN: \nStrictly used when a user requests to open, show, display, or fetch a chart, report, dashboard, or table. The intent is to open or navigate to a predefined object (chart, report, view or dashboard).\nOPEN actions are typically standalone initial queries (not follow-ups).\nIf an \"open\" action query is preceded by an analytical query, treat it as a new intent unless contextually overridden. \nopen <view/chart/report/table name>\nopen the <view/chart/report> chart\nshow me the <view/chart/report> report\n\nExamples: \n\"show me the revenue report\" -> skill: \"ACTION\", action: \"OPEN\", sub_skill: \"REPORT\", \"is_follow_up\":false\n\"display the customer dashboard\" -> skill: \"ACTION\", action: \"OPEN\", sub_skill: \"DASHBOARD\", \"is_follow_up\":false\n\"open Churn Rate by Year - recommendation\" -> skill: \"ACTION\", action: \"OPEN\", sub_skill: \"REPORT\", \"is_follow_up\":false\n\"open view Churn Rate by Year - recommendation\" -> skill: \"ACTION\", action: \"OPEN\", sub_skill: \"REPORT\", \"is_follow_up\":false\n\n\nPrevious query: sales across headquarters\nCurrent query: open sales by regions chart -> \"skill\":\"ACTION\", action: \"OPEN\", sub_skill: \"REPORT\", \"is_follow_up\":false\n\nDELETE: \nIf the user requests to delete a entity.\n\nSAVE: \nIf the user requests to save a chart,report,view, dashboard, table or query table\n\nLIST: \nIf the user wants to list or get any chart, report, view, dashboard, table or query table.Note : STRICTLY Always classify ONLY \'get\' and \'list\' actions explicitly as \'LIST\' action\n\n### Rules:\n1. Query Follow-Up (follow_up_type: \"1\"):\n\nSet to \"is_follow_up\" as true and \"follow_up_type\" as \"1\" if the query builds upon a previous user query (e.g., refining or continuing a request).\nExample:\nPrevious Query: \"Show me the revenue report.\" -> skill: \"ACTION\", sub_skill: \"OPEN\", sub_skill: \"REPORT\"\nCurrent Query: \"Save this report as PDF.\" -> \"is_follow_up\": true, \"follow_up_type\": \"1\"\n\n2. Output Follow-Up (follow_type: \"2\"):\n\nSet to \"is_follow_up\" as true and \"follow_up_type\" as \"2\" if the query acts upon the previous output (e.g., applying an action to retrieved data).\nExample:\nPrevious Query: \"create a dashboard with sales metrics\" -> skill: \"ACTION\", sub_skill:\"DASHBOARD\", action: \"CREATE\"\nCurrent Query: \"share this dashboard.\" -> \"is_follow_up\": true, \"follow_up_type\": \"2\", \"sub_skill\":\"DASHBOARD\"\n\nA query should be classified as an output follow-up (is_output_follow_up: true) if it refers to a specific item retrieved in the previous response (e.g., applying an action to listed reports, charts, or dashboards)\n3. If a user modifies an existing chart within an analytical context (e.g., changing visualization type, filtering data, or refining analysis), it should remain classified under ANALYTICAL -> ZIA and marked as an output follow-up (follow_up_type: \"2\")\n\n4. Chart Modifications (is_query_follow_up: true, still under ZIA)\nRequests like \"Change the above chart to a bar chart\" should not be reclassified as ACTION -> MODIFY.\nInstead, treat them as analytical refinements, staying within ANALYTICAL -> ZIA.\n\n5. - **Populate \"action\" key only when skill = \"ACTION\". While picking the action, make sure it is from list of supported actions, otherwise do not classify it as ACTION skill.\n6. Adhere strictly to the predefined set of skills, sub-skills, and associated actions as specified in the configured scenarios. However, when the current query appears to be a continuation of the previous assistant output or aligns with the follow-up rules, prioritize it as a valid follow-up before deciding to fall back to FAQ or CHITCHAT. Only redirect to FAQ or CHITCHAT if follow-up classification fails and no valid skill/sub-skill/action mapping exists.\n\n7.In the output JSON, if the values are empty do not populate the key in the results.\n8.Always prioritise the follow-up prompt rules, before falling back to FAQ or CHITCHAT skills\n9. If the assistant previously requested an input value (via \"Client Output\", e.g., \"Please provide the name\"), treat the next user query as a contextual response even if it is a single word or phrase. In such cases:\n- Set `is_follow_up: true`, `follow_up_type: \"2\"`, with skills and sub_skills populated appropriately\n- Do not assign the query to CHITCHAT or FAQ skills.\n- Retain the previous skill/sub-skill/ from the context, with action as \"FORM\"\n- Populate the appropriate key in the actual JSON based on what was asked in the Client Output.\n- If the assistant response was a prompt to provide a name, title, or other field (e.g., \"provide the title\", \"enter a name to save\"), then treat **any non-command input** (including phrases like \"test1\", \"sales across regions\", or \"kkkk\") as a **follow-up to the current FORM action**. These may resemble queries for other skills like ZIA or CHITCHAT, but should **not break context** as long as they are part of a form-filling or continuation step.\n-Only treat the query as a **new instruction** if it clearly initiates a different action unrelated to the form request (e.g., \"create chart\", \"list dashboards\")\n10. If the user responds casually with messages like \"okay\", \"good\", or similar acknowledgments after the assistant\'s response, do not classify the query under the \"ACTION\" skill or the \"FORM\" sub_skill. Instead, reclassify such responses appropriately as \"CHITCHAT\".\n11. For queries with any actions like email this, schedule this, export this, etc as a follow up to any valid previous queries/outputs must be tagged as follow-up.\n\nPopulate the key \"follow_up_type\" with the value \"1\" or \"2\", based on the type of follow-up.\n\nYou receive a list of user and assistant interactions. Each user query is prefixed with \"Q1\", \"Q2\", etc., and is followed by the assistant’s processed output. Your task is to understand each user query and its corresponding assistant response, and generate a final structured JSON in the specified in the \"Output JSON Format\" section.\n\nExample Input Format:\n[{\"role\":\"user\",\"content\":\"Q1, Q2, Q3,…\"},{\"role\":\"assistant\",\"content\":\"<processed_structured_output>\"},{\"role\":\"user\",\"content\":\"<current_user_query>\"}]\n\nGuardrail Rules:\nIf the user query involves any of the following combinations of action and entity, mark the query as **ACTION** skill without specifying the sub-skills:\n- Create / Add / Rename: Table, Column, Alert, User, User group\n- Open: Zia Settings, Org Settings, Workspace Settings, Model Diagram, Data Model\n- Share to: Slack, Cliq, Support, Publish\n- Delete / Revoke / Edit: Schedulers, Shared permissions, Lookups, Columns, Tables, Reports, Alerts, Users, User roles, User permissions\n- Schedule: Sync\n- List: Email schedules, Users, Alerts, Columns, Shared views, Variables, Sync schedules\n\nFor these unsupported combinations:\n- Do **not** assign any sub-skill (e.g., QUERYTABLE, DASHBOARD, SLIDESHOW, ZIA,.etc)\n- Set only the skill to \"ACTION\"\n\nFor every query, return the classification in the following strict JSON format. Do not add any additional explanation.\nStrictly return only this JSON object. Do not return OpenAI-style role-based messages or intermediate analytical structures. Do not generate filters, aggregates, or grouping instructions. \n\nSTRICTLY ENSURE ONLY THESE KEYS \"skill, action, sub_skill, selected_values, is_follow_up, follow_up_type, response\" ARE PRESENT IN YOUR OUTPUT JSON. Only return a single JSON object with the following fields:
za.askzia.generic_classifier.action.followup.prompt=Determine if the current query is a continuation of the previous skill. If it is, identify if it is a follow-up at the query level or output level. If it is a follow-up at the query level, set `is_follow_up: true, follow_up_type:\"1\"`. If it is a follow-up at the output level, set `is_follow_up: true, follow_up_type:\"2\"`. If the query is independent, set `is_follow_up: false`.\n\nLevel 1 Classification (Json Template Based - Query Follow-Up):\n\n- If the query logically continues from the previous query\'s outcome, mark it as `is_follow_up: true, follow_up_type:\"1\"`.\n- If the query modifies (add\\/remove\\/replace\\/filter) the keys in the Json template, mark it as `is_follow_up: true, follow_up_type:\"1\"`.\n\nLevel 2 Classification (Output Based & Supported Actions Based - Output Follow-Up):\n\n- If the user asks questions about the output, mark it as `is_follow_up: true, follow_up_type:\"2\"`.\n- If the user requests actions on the output, mark it as `is_follow_up: true, follow_up_type:\"2\"`. (Examples: share this dashboard, export as image, export as pdf,.etc)\n- If the user requests the same actions without involving the previous output, consider it a new query (`is_follow_up: false`).\n- For the \"share\" action, mark it as `is_follow_up: true, follow_up_type:\"2\"`, while maintaining the previous module information\n- Pick the \"Output\" values given to populate the `selected_values` key if the user specifies any in the input query. Strictly pick numbers only from the provided \"Output\" list.If the user:\n- Repeats the same intent with a new sub_skill, without specifying explicitly (e.g., from \"export sales charts\" -> \"I want dashboards\")\n- Provides a short phrase or sub_skill name (eg. I want dashboards)\nthen treat it as an output follow-up (`follow_up_type: \"2\"`), continuing the last known skill\\/action\\/sub_skill unless the user clearly changes it.\nExample:\nInput:\n[\n  { \"role\": \"user\", \"content\": \"export sales charts\" },\n  { \"role\": \"assistant\", \"content\": \"List sales charts to export\" },\n  { \"role\": \"user\", \"content\": \"I want dashboards\" }\n]\nOutput:\n{\n  \"skill\": \"ACTION\",\n  \"sub_skill\": \"DASHBOARD\",\n  \"action\": \"EXPORT\",\n  \"is_follow_up\": true,\n  \"follow_up_type\": \"2\"\n}\nExplanation: Here the user intent is to get the list of dashboards in place of previous intent with charts. Only sub_skill need to be modified, by maintaining the same skill and action.\n\nStrictly enforcement rules:\n - The follow_up_type must be either \"1\" or \"2\", based on the detected follow-up type. Both values must not be set simultaneously. This validation must be performed rigorously before generating the results.\n - If \"Client Output\" is present in the assistant output, consider the following query as a contextual response to the assistant output. For these cases, set `is_follow_up: true, follow_up_type:\"2\"` considering the context and the current user query.\n\nExamples:\n\nScenario 1:\n\nJson Template: {\"reportname\":\"\",\"reportfolder\":\"\",\"dashboardname\":\"\",\"exporttype\":\"CSV|PDF|TEXT|IMG\",\"multiplematchesinput\":\"\"}\nActual Json: {\"reportname\":\"sales\",\"exporttype\":\"IMG\"}\nOutput: There are multiple matches present, please select a match: {\"1\":\"Sales Pie\",\"2\":\"sales by status\",\"3\":\"status-wise sales\",\"4\":\"sales Across Years\",\"5\":\"ship mode-wise Sales\"}\n\nInput: create a slideshow -> Output: {\"skill\":\"Action\",\"is_follow_up\":false}\nInput: create a dashboard -> Output: {\"skill\":\"Action\",\"is_follow_up\":false}\nInput: total profit for 2023 -> Output: {\"skill\":\"Analytical\",\"is_follow_up\":false}\nInput: share sales by region chart -> Output: {\"skill\":\"Action\",\"is_follow_up\":false}\nInput: export profit chart -> Output: {\"skill\":\"Action\",\"is_follow_up\":false}\nInput: change the report name to \"sales by region\" -> Output: {\"skill\":\"Action\",\"\"is_follow_up\":true, \"follow_up_type\":\"1\"}\nInput: as csv -> Output: {\"skill\":\"Action\",\"\"is_follow_up\":true, \"follow_up_type\":\"1\"}\nInput: share now -> Output: {\"skill\":\"Action\",\"is_follow_up\":true, \"follow_up_type\":\"2\"}\nInput: share the ABOVE to Sandeep -> Output: {\"skill\":\"Action\",\"is_follow_up\":true, \"follow_up_type\":\"2\",\"action\":\"share\"}\nInput: select 1st and 4th ones -> Output: {\"skill\":\"Action\",\"is_follow_up\":true, \"follow_up_type\":\"2\",\"selected_values\":[\"1\",\"4\"]}\nInput: export the 1st one -> Output: {\"skill\":\"Action\",\"is_follow_up\":true, \"follow_up_type\":\"2\",\"selected_values\":[\"1\"],\"action\":\"export\"}\nInput: share all the ABOVE results to Sandeep -> Output: {\"skill\":\"Action\",\"is_follow_up\":true, \"follow_up_type\":\"2\",\"selected_values\":[\"1\",\"2\",\"3\",\"4\",\"5\"],\"action\":\"share\"}\nInput: now create a mail schedule to send this report\\/dashboard with John and Lisa every Friday at 3pm -> Output: {\"skill\":\"Action\",\"is_follow_up\":false, \"follow_up_type\":\"2\",\"action\":\"create\"}\n\nScenario 2:\n\nJson Template: {\"reportname\":\"\",\"chartname\":\"\",\"apply_filter\":\"\",\"customize\":\"\",\"multiplematchesinput\":\"\"}\nActual Json: {\"chartname\":\"sales by region\"}\nOutput: There are multiple matches present, please select a match: {\"1\":\"Sales Pie\",\"2\":\"sales by status\",\"3\":\"status-wise sales\",\"4\":\"sales Across Years\",\"5\":\"ship mode-wise Sales\"}\n\nInput: apply filter for last 3 months -> Output: {\"skill\":\"Action\",\"\"is_follow_up\":false, \"follow_up_type\":\"1\",\"action\":\"filter\"}\nInput: change color theme -> Output: {\"skill\":\"Action\",\"\"is_follow_up\":false, \"follow_up_type\":\"1\",\"action\":\"modify\"}\nInput: highlight regions with highest sales -> Output: {\"skill\":\"Action\",\"\"is_follow_up\":true, \"follow_up_type\":\"1\"}\nInput: select the 2nd and 5th ones -> Output: {\"skill\":\"Action\",\"is_follow_up\":true, \"follow_up_type\":\"2\",\"selected_values\":[\"2\",\"5\"]}\n\nScenario 3:\n\nJson Template: {\"tablename\":\"\",\"rowname\":\"\",\"columnname\":\"\",\"operation\":\"\",\"multiplematchesinput\":\"\"}\nActual Json: {\"tablename\":\"sales_data\"}\nOutput: There are multiple matches present, please select a match: {\"1\":\"Region-wise Sales\",\"2\":\"Monthly Sales\",\"3\":\"Yearly Sales\",\"4\":\"Product-wise Sales\",\"5\":\"Customer-wise Sales\"}\n\nInput: show total revenue -> Output: {\"skill\":\"Analytical\",\"is_follow_up\":false, \"follow_up_type\":\"1\"}\nInput: filter data for Q1 2024 -> Output: {\"skill\":\"Action\",\"is_follow_up\":true, \"follow_up_type\":\"1\"}\nInput: sort by highest revenue -> Output: {\"skill\":\"Action\",\"is_follow_up\":true, \"follow_up_type\":\"1\"}\nInput: select the 3rd and 4th ones -> Output: {\"skill\":\"Action\",\"is_follow_up\":true, \"follow_up_type\":\"2\",\"selected_values\":[\"3\",\"4\"]}\n\nScenario 4:\nJson Template: {\"dashboardname\":\"\",\"viewname\":\"\",\"viewtype\":\"\",\"reportname\":\"\",\"tablename\":\"\",\"permission\":\"\",\"insights\":\"\",\"slideshowtheme\":\"\",\"slideshowname\":\"\",\"multiplematchesinput\":\"\"}\nActual Json: {\"slideshowname\":\"Slideshow Presentation Overview\"}\nClient Output: Please provide the viewname\"\nThe action needs to be picked based on the previous query, while considering the current query as a follow up.\nInput: sales -> Output: {\"skill\":\"Action\",\"\"is_follow_up\":true,\"action\":\"FORM\"}\nInput: test slideshow -> Output: {\"skill\":\"Action\",\"is_follow_up\":true,\"action\":\"FORM\"}
za.askzia.generic_classifier.followup.prompt=The previous query is processed as \"ZIA\" skill. Detect whether the current user query is a continuation, modification, or refinement of a previous context.\r\n\r\nBelow are the rules to determine whether it is a follow-up or not, it the query:\r\n1. Entity Modification: maintains the same analytical context but changes the entities dimension or measure, it is a follow-up. (e.g., replace sales with profit, remove sales)\r\n2. Time Frame Change: modifies the time frame of the previous query (e.g., Q1 to Q2), it is a follow-up.\r\n3. Filter Adjustment: adds, modifies, or removes filters on the same analysis (e.g., categories, ship modes), it is a follow-up.\r\n4. Visualization Modification: changes the chart type or visualization elements (e.g., switch to pie chart, modify tooltip), it is a follow-up.\r\n5. Drill-Down Request: requests a deeper drill-down (e.g., from category to subcategory), it is a follow-up.\r\n6. Forecast\/Trend Continuation:continues a forecasting or trend-based analysis, it is a follow-up.\r\n7. Contextual Additions: adds clarifications or asks for additional data in the same context (e.g., include discounts), it is a follow-up.\r\n8. Implicit References: refers to the previous query using pronouns or implicit references (e.g., \"what about\", \"include that\"), it is a follow-up.\r\n9. New Analytical Topic: switches to a different analytical topic or intent involving unrelated entities(new metric, dimension), it is not a follow-up.\r\n10. Context Reset or Chitchat: resets context or shifts to a different domain like chitchat, it is not a follow-up.\r\n11.  Analytical Computation on Prior Output: If the query applies analytical computations or statistical methods such as correlation, contribution, anomaly detection, etc., on top of a previously referenced analysis or dataset or output (even implicitly), it must be classified as a follow-up.\r\n12. For the follow-up actions on top of the opened chart\/views or generated views, like exporting the result as Image, pdf or csv, exporting this view, schedule this view,. etc, strictly tag it as `is_follow_up: true`\r\n13.If the current query appears to be independent or unrelated to its previous queries, strictly do not classify it as a follow-up.\r\n14. Strict Logical Connection: If the current query does not explicitly reference or logically connect to the previous query (e.g., no shared entities, dimensions, or context), classify it as not a follow-up, even if it appears similar in structure or intent. For example:\r\nq1) sales q2) cost Output: \"is_follow_up\": false\r\nFor the follow-up cases, mark it as `is_follow_up: true` in the output JSON\r\nFor unrelated or independent queries, include \"is_follow_up\": false.\r\nFew shot examples using below SQL Table structures:\r\n\r\nSQL Table structures: \r\n{Cluster 1: [Order(\"Order ID\" int,\"Product ID\" int,\"Customer ID\" int,\"Order Date\" date,\"Order Priority\" text,\"Order Quantity\" int,\"unit price\" int,\"profit\" int,\"Sales\" int,\"Discount\" int,\"ship mode\" text,\"shipping cost\" int),\r\nCustomers(\"Customer ID\" int,\"Customer Name\" text,\"Province\" text,\"Region\" text,\"Customer Segment\" text),\r\nProducts(\"Product ID\" int,\"Product Name\" text,\"Product Subcategory\" text,\"category\" text,\"Container\" text,\"Unit Price\" int)],\r\n\r\nCluster 2:\r\n[Address(\"ID\" text,\"Address Line1\" text,\"City\" text,\"District\" text,\"State\" text,\"Postal Code\" text),\r\nCart(\"Sales Order ID\" int,\"Product ID\" int,\"Quantity\" int),\r\nCompany(\"Company-ID\" int,\"Name\" text,\"Head Quarters\" text,\"Revenue\" int),\r\nCustomer(\"Customer-ID\" int,\"Company Name\" text,\"Contact Person\" text,\"Email\" text,\"Phone 1\" int,\"Billing Address\" text,\"Shipping Address\" text,\"FirstName\" text,\"LastName\" text),\r\nProduct(\"ProductID\" int,\"Name\" text,\"Type\" text,\"Cost\" int,\"Price\" int,\"Modified Date\" date),\r\nSales Order(\"Sales_Order_ID\" int,\"userid\" int,\"order-date\" date,\"status\" text,\"sales\" int,\"profit\" int,\"loss\" int,\"Customer ID\" int),\r\nUser(\"User__ID\" int,\"Company_ID\" int,\"First Name\" text,\"Last Name\" text,\"Email Address\" text,\"Date Created\" date,\"Last Login\" date,\"User Type\" text,\"Address ID\" text)]}\r\n\r\nExamples for \"ZIA\" skill follow-ups in various scenarios:\r\n\r\nExample 1: Same cluster, logical follow-up\r\nInput 1: Create a chart for sales by region\r\nInput 2: Now show me the profit by product name.\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": true, \"follow_up_type\":\"1\"}\r\n\r\nExample 2: Same cluster, logical follow-up\r\nInput 1: Sales by region for Q1 2010.\r\nInput 2: What about for Q2 2010?\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": true, \"follow_up_type\":\"1\"}\r\n\r\nExample 3: Different cluster, not a follow-up\r\nInput 1: Show me the sales by product name.\r\nInput 2: Get me company names based on revenue.\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": false}\r\n\r\nExample 4: Same cluster, logical follow-up, with axis modifications \r\nInput 1: List the products with quantity greater than 1500.\r\nInput 2: And what about for products in the \"Furniture\" category?\r\nInput 3: Put the product name in the tool tip\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": true, \"follow_up_type\":\"1\"}\r\n\r\nExample 5: Different cluster, not a follow-up\r\nInput 1: List all regions with the highest number of customers.\r\nInput 2: Show me the addresses by state.\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": false}\r\n\r\nExample 6: Follow-up changing filters\r\nInput 1: Show me the sales by product category.\r\nInput 2: How about adding the ship mode to the breakdown?\r\nInput 3: remove delivery truck ship mode from analysis\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": true, \"follow_up_type\":\"1\"}\r\n\r\nExample 7: Follow-up clarification\r\nInput 1: Get me total sales for the last three years.\r\nInput 2: Include the discounts too.\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": true, \"follow_up_type\":\"1\"}\r\n\r\nExample 8: Different cluster, standalone request\r\nInput 1: Show me the average profit by shipping mode.\r\nInput 2: What is the total revenue for each company?\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": false}\r\n\r\nExample 9: Same cluster, follow-up with chart type\r\nInput 1: Show me the sales by region.\r\nInput 2: Can you display this as a pie chart?\r\nInput 3: convert it to a line chart\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": true, \"follow_up_type\":\"1\"}\r\n\r\nExample 10: Clear context reset with \"CHITCHAT\"\r\nInput 1: Show me sales in India.\r\nInput 2: Clear history.\r\nOutput: {\"skill\": \"CHITCHAT\"}\r\n\r\nExample 11: Follow-up with additional condition\r\nInput 1: Get me the sales by product name in 2010.\r\nInput 2: What about for 2011?\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": true, \"follow_up_type\":\"1\"}\r\n\r\nExample 12: Different cluster, unrelated question\r\nInput 1: List the customer names by region.\r\nInput 2: What is the postal code for New York?\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": false}\r\n\r\nExample 13: Follow-up with deeper drill-down\r\nInput 1: Get me the sales by product category.\r\nInput 2: Can you break this down further by product subcategory?\r\nInput 3: Put the date column in colour axis\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": true, \"follow_up_type\":\"1\"}\r\n\r\nExample 14: Different cluster, unrelated aggregation\r\nInput 1: Show me the revenue by company headquarters.\r\nInput 2: What is the total profit by shipping mode?\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": false}\r\n\r\nExample 15: Follow-up continuation\r\nInput 1: Show me the sales trend over the last five years.\r\nInput 2: Can you also include the profit trend?\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": true, \"follow_up_type\":\"1\"}\r\n\r\nExample 16: Follow-up continuation using a concept\r\nInput 1: give me the order forecast\r\nInput 2: forecast monthly orders count for next 5 months\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": true, \"follow_up_type\":\"1\"}\r\n\r\nExample 17: Implicit Follow ups\r\nInput 1: sales across years\r\nInput 2: show this by months\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": true, \"follow_up_type\":\"1\"}\r\n\r\nExample 18 : Insights query\r\nInput 1: sales across years\r\nInput 2: get me the insights\r\nInput 3: sales\r\nOutput: {\"skill\": \"ANALYTICAL\", \"sub_skill\": \"ZIA\", \"is_follow_up\": false}\r\n
za.askzia.generic_classifier.gptsql.followup.prompt=The previous query is processed as \"GPT-SQL\" skill. Your goal is to evaluate whether the user intends to modify, refine, or extend the previously generated SQL query even when the current query is brief, implicit, or consists of a single word. These follow-up inputs may include actions such as adding or removing columns, applying filters, adjusting date ranges, modifying sort orders, or limiting results.\n\nFor instance, follow-ups like \"add region\", \"sort descending\", \"last year\", or \"limit to top 10\" should be treated as extensions of the previous SQL query context. These inputs might not explicitly mention SQL but are semantically intended to alter or enhance the last generated SQL output.\n\nIf the current query introduces a completely different intent such as asking for a new SQL query on an unrelated topic, requesting a definition or explanation, or shifting the task away from SQL generation then it should not be considered a follow-up.\n\nYou must analyze the semantic connection between the current query and the previous one to determine whether it continues the context of SQL generation.\n\nPositive Examples  Mark it as  `skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\", is_follow_up: true, follow_up_type:\"1\"` in the output JSON\n- \"add region\"\n- \"sort it by total profit\"\n- \"only last month\"\n- \"filter for USA\"\n- \"include order date\"\n- \"limit to top 10\"\n- \"now by category\"\n- \"make it descending\"\n- \"also show discount\"\n\nNegative Examples  Mark it as ``skill: \"ANALYTICAL\", sub_skill: \"GPT-SQL\", is_follow_up: false` in the output JSON\n- \"generate a SQL query for sales by region\" (new intent)\n- \"explain how SQL joins work\"\n- \"what is a left join\"\n- \"create table for customers\"\n- \"how to optimize this query\"\n- \"clear this\"\n- \"show me query for a different report\"
za.askzia.generic_classifier.gptformula.followup.prompt=The previous query is processed as \"GPT-FORMULA\" skill. Detect whether the current user query is a continuation, modification, or refinement of a previously generated custom or aggregate formula. This includes both explicit and implicit follow-ups.\n\nA query should be treated as a follow-up under GPT-FORMULA if:\n- It refers to modifying the previous formula (e.g., \"add discount,\" \"remove profit,\" \"make it conditional,\" \"change sum to average\").\n- It includes vague or brief instructions that rely on the prior context (e.g., \"make it percentage,\" \"add filter,\" \"group it,\" \"only for this year\").\n- It introduces a new requirement that modifies an earlier formula but does not restate the entire logic, implying that the original formula should be retained or partially reused.\n- It appears independent on the surface, but shares the same intent, context, column references, or aggregation structure as the immediate past query and is meant to extend or adjust the logic of the previously generated expression.\n\nFor these cases, mark it as `is_follow_up: true, follow_up_type:\"1\"` in the output JSON\n\nThe detection should be sensitive to the semantic continuity between the previous and current queries, even when the user does not explicitly mention that it is a follow-up.\nDo not mark the query as a follow-up if:\n- It initiates a clearly different skill (e.g., switches to SQL SELECT generation or chart generation).\n- It starts with a new calculation intent unrelated to the previous one (e.g., different subject, columns, or metric).\n\nFor these cases, mark it as `is_follow_up: false` in the output JSON
za.askzia.generic_classifier.slideshow.followup.prompt=Determine whether the current query is a continuation of the \"SLIDESHOW\" skill. If it is, classify the follow-up type appropriately:\n1. Set \"is_follow_up\": true, \"follow_up_type\": \"1\" for query-level follow-ups, such as modifying parameters (theme, ordering, slide title).\n2. Set \"is_follow_up\": true, \"follow_up_type\": \"2\" for output-level follow-ups, such as acting on client outputs, selected slides, or previously displayed options.\n3. Set \"is_follow_up\": false for new, independent queries.\nDetermine whether the current user query is a follow-up or not based on the \"Client Output\" present in previous output, if it is not empty. Otherwise determine based on other parameters and context.\nClassification Rules:\n\nLevel 1  Query Follow-Up (follow_up_type: \"1\", action:\"FORM\"):\n1. If the query modifies slideshow metadata (e.g., title, theme, layout, order, transitions).\n2. If the query applies filters or updates to an existing or previously mentioned slideshow.\n3. If the query refines the current slideshow context without requiring a new action or unrelated module.\n\nLevel 2  Output Follow-Up (follow_up_type: \"2\", action:\"FORM\"):\n1. If the query responds to a system prompt asking for a missing detail (e.g., \"Please provide the viewname\").\n2. If the query selects slides, views, or applies actions on a listed output (e.g., Select the 1st and 3rd views).\n3. If the query requests to share\\/export\\/update based on output shown by the assistant.\n\nOnly use \"action\": \"CREATE\" when the user is creating a slideshow.All other operations related to slides, views, themes, filters, transitions, etc. like ADD, DELETE, MODIFY, REORDERING should be strictly classified as \"action\": \"MODIFY\", since \"MODIFY\" action all these actions internally.\n\nFollow-Up Logic:\n1. \"is_follow_up\": true, \"follow_up_type\": \"1\"When user query:\n    * Modifies slideshow metadata (title, theme, order, transitions)\n    * Applies filters or updates to an existing\\/mentioned slideshow\n    * Refines ongoing slideshow context.\n2. \"is_follow_up\": true, \"follow_up_type\": \"2\"When user query:\n    * Responds to assistant prompt (e.g., Please provide the name)\n    * Acts on displayed output (e.g., select 2nd view, share this slideshow)\n3. \"is_follow_up\": falseWhen query starts a new task, unrelated to slideshow context.\nNever classify form completions (e.g., view names, titles) as CHITCHAT.\n\nStrict Rules & Exclusions:\nDo not mark prompt-based form inputs (e.g., slide name, view name, text inputs) as follow-ups. These should be treated as part of form completion and not tagged as CHITCHAT or follow-ups.\n1. Only tag a query as follow-up if it modifies or acts on an already established slideshow context.\n2. follow_up_type must be strictly one of \"1\" or \"2\" - never both.\n3. If \"Client Output\" is present and the query logically responds to it, mark it as follow_up_type: \"2\".\n\nIf the assistant response was a prompt to provide a name, title, or other field (e.g., provide the title, enter a name to save), then treat **any non-command input** (including phrases like \"test1\", \"sales across regions\", or \"kkkk\") as a **follow-up to the current FORM action**. These may resemble queries for other skills like ZIA or CHITCHAT, but should **not break context** as long as they are part of a form-filling or continuation step.\n-Only treat the query as a **new instruction** if it clearly initiates a different action unrelated to the form request (e.g., \"create chart\", \"list dashboards\")\n\n### Example 1\nPrevious Query: \"create a slideshow\"\nAssistant Response: \"provide the name\"\nCurrent Query: \"data science\"\nExpected Output:\n{\n  \"skill\": \"ACTION\",\n  \"sub_skill\": \"SLIDESHOW\",\n  \"Action\": \"FORM\",\n  \"is_follow_up\": true,\n  \"follow_up_type\": \"1\",\n  \"response\": \"\"\n}\n\n### Example 2\nCurrent Query: \"How do I create a slideshow?\"\nExpected Output:\n{\n  \"skill\": \"FAQ\",\n  \"sub_skill\": \"\",\n  \"is_follow_up\": false,\n  \"follow_up_type\": \"0\",\n  \"response\": \"\"\n}\n\n### Example 3\nCurrent Query: \"Clear history\"\nExpected Output:\n{\n  \"skill\": \"CHITCHAT\",\n  \"sub_skill\": \"\",\n  \"is_follow_up\": false,\n  \"follow_up_type\": \"0\",\n  \"response\": \"Sure, your conversation history has been cleared.\"\n}\n\n### Example 4\nPrevious Query: \"create a slideshow\"\nAssistant Response: \"Slideshow Created\"\nCurrent Query: \"add a slide at the end\"\nExpected Output:\n{\n  \"skill\": \"ACTION\",\n  \"sub_skill\": \"SLIDESHOW\",\n  \"Action\": \"MODIFY\",\n  \"is_follow_up\": true,\n  \"follow_up_type\": \"2\",\n  \"response\": \"\"\n}\n\n### Example 5\nPrevious Query: \"create a slideshow\"\nAssistant Response: \"Slideshow Created\"\nCurrent Query: \"delete the 2nd slide\"\nExpected Output:\n{\n  \"skill\": \"ACTION\",\n  \"sub_skill\": \"SLIDESHOW\",\n  \"Action\": \"MODIFY\",\n  \"is_follow_up\": true,\n  \"follow_up_type\": \"2\",\n  \"response\": \"\"\n}\n\n### Example 6:\nPrevious Query: \"create a slideshow using sales views\"\nAssistant Response: \"Slideshow Created\"\nCurrent Query: \"change it without login\"\nExpected Output:\n{\n  \"skill\": \"ACTION\",\n  \"sub_skill\": \"SLIDESHOW\",\n  \"Action\": \"MODIFY\",\n  \"is_follow_up\": true,\n  \"follow_up_type\": \"2\",\n  \"response\": \"\"\n}\n\n### Example 7\nPrevious Query: \"create a slideshow\"\nAssistant Response: \"provide the title\"\nCurrent Query: \"sales and cost across years for each product category\"\nExpected Output:\n{\n  \"skill\": \"ACTION\",\n  \"sub_skill\": \"SLIDESHOW\",\n  \"Action\": \"FORM\",\n  \"is_follow_up\": true,\n  \"follow_up_type\": \"1\",\n  \"response\": \"\"\n}\n\n### Example 8\nPrevious Query: \"rename the slideshow\"\nAssistant Response: \"provide the title\"\nCurrent Query: \"kkkkkk\"\nExpected Output:\n{\n  \"skill\": \"ACTION\",\n  \"sub_skill\": \"SLIDESHOW\",\n  \"Action\": \"FORM\",\n  \"is_follow_up\": true,\n  \"follow_up_type\": \"1\",\n  \"response\": \"\"\n}
za.askzia.generic_classifier.dashboard.followup.prompt=Dashboard Show - Followup Prompt\n\nDetermine whether the current query is a continuation of the \"DASHBOARD\" skill. If it is, classify the follow-up type appropriately:\n1. Set \"is_follow_up\": true, \"follow_up_type\": \"1\" for query-level follow-ups, such as modifying parameters (theme, ordering, view/dashboard title, add/remove views/cards/widgets)\n2. Set \"is_follow_up\": true, \"follow_up_type\": \"2\" for output-level follow-ups, such as acting on client outputs, selected views, or previously displayed options.\n3. Set \"is_follow_up\": false for new, independent queries.\nUse the following logic to classify the query:\n A query should be classified as a follow-up if:\n1. It does not mention creating a new dashboard but refers to modifying or interacting with a dashboard.\n2. It omits the dashboard name but uses phrases like:\n    1. -\"add this\", \"resize this\", \"move this to the left/right\", \"change the theme\", \"apply dark mode\", \"rearrange the views\", \"update the layout\", \"make it compact\", etc.\n3. It uses contextual references such as: \"this view\", \"the dashboard\", \"current dashboard\", \"the previous layout\", \"that widget\"\n4. It involves layout, visual, or theme adjustments after a known creation step (e.g., resizing views, changing theme colors).\n5. It implies continuation or enhancement of a recently created dashboard.\n6. Add/remove charts/widgets/reports/views present in the dashboard. The chart/view/report names used to modify the operations might represent some other skills sometimes, but stick to DASHBOARD skill. Strictly do not go out of the DASHBOARD skill scope.\n7. When user tries to export/share the dashboard like \"share the above dashboard\", \"export as pdf\" etc. it needs to be treated as follow-up.\n8.For any actions that is performed on top of the dashboard, like add/remove/modify - views/widgets/reports/cards/charts, it should be strictly tagged as \"is_follow_up\": true and \"action\": \"MODIFY\", with sub_skill as DASHBOARD. Strictly do not tag it as REPORT or any other skill.\n Examples for this scenario are:\n* Remove empty widgets\n* Clean up reports that return blank or null results\n* Exclude charts without any data\n* Hide visualizations with zero records\n* Discard tiles showing no information\n* Filter out non-populated reports\n* Remove insights where data is missing\n* Eliminate KPIs or charts without values\n* Delete summary views that have no entries\n* remove sales reports\n* remove deals pipeline by amount reprt\n* add leads by source monthly trend chart\n* Delete reports with no data\n* Delete reports with no data\n\nA query is not a follow-up if:\n1. It clearly states an intent to create a new dashboard with a name, e.g., \"Create a dashboard named SalesTrends\".\n2. It starts from scratch without assuming prior context or without referencing existing dashboards.\n\nExamples:\nSet as  \"is_follow_up\": true, \"follow_up_type\": \"2\"\n* Add this pie chart to the dashboard \n* Remove empty widgets from the dashboard\n* Remove empty widgets\n* Move the revenue chart to the first row \n* remove total sales across each country report\n* remove all sales related charts\n* remove all reports involving profit\n* Set the shadow color of the dashboard widgets to grey. \n* Change the grid color of the cards to light blue. \n* Merge the overview with the dashboard \n* Shuffle the views\n* Organize the cards in my dashboard differently \n* Change the size of the chart\n* Diminish the width of the view.\n* Compress the view. \n* Change theme\n* Suggest me few themes.\n* Change the card colour to black\n* Remove sales across region\n* remove total sales across each quarter\n* Export as csv\n* Export as html\n* Export as pdf\n\nSet as  \"is_follow_up\": false\n* Create a dashboard named RegionalSales with three KPI widgets\n* Create a new dashboard
INITIAL_SUGGESTED_QUERIES_XML_FILE=./conf/nlp/dictionary/RAZ/QuerySuggestion/query_suggestion_template.xml

FOLLOWUP_QUERIES_XML_FILE=./conf/nlp/dictionary/RAZ/FollowupQueries/followup_query_pattern.xml

za.slideshow.modify.askzia.prompt.for.zia.turbo=Task Overview: You are an AI assistant tasked with modifying a slideshow's details based on user queries. The slideshow data is provided as a JSON object, containing all relevant information about the slideshow and its slides.\n\nInput Format:\n- Input Slideshow JSON Object: This object includes metadata and detailed information about the slideshow, such as details about each slide.\n- User Query: A natural language request from the user specifying the modification needed for the slideshow.\n\nSlideshow JSON Object Description:\n- presentationInformation:\n  - title: A brief and descriptive name summarizing the slideshow's content or purpose. Also called as slideshow title\n  - theme: The visual theme, including color schemes, fonts, and design elements for a cohesive look. supported themes : "WHITEPAPER, BEEHIVE, EXPERIMENTAL, STROKES, THEATRICAL"\n  - withlogin: This indicates whether slideshow is accessible with login or not. If it is false, it means the slideshow is made public.\n\n- slidesInformation: An array with detailed information about each slide. Each object represents a single slide and includes:\n  - title: A succinct description of the slide's content or focus. This is title of that particular slide.\n  - slideId: A unique identifier for the slide, ensuring sequential order.\n  - viewid: An array of objects representing views associated with the slide, including metadata like title, id, type, and reporttype.\n\nInstructions:\n- Read the provided slideshow JSON object and user query.\n- Identify the specific modifications required based on the user query.\n- Provide only the modified, added, or deleted information of keys and values in a structured format.\n- If the user provides partial input or if clarification is needed, ask a question in the format: `{"question":""}`. Ensure all questions starts with Caps.\n- When adding a new slide, If no position is specified, add the new slide as the last slide. For a new slide addition.\n- Strictly give proper json as output. Also, validate whether the given output is proper Json format.\n- These are the supported themes : "WHITEPAPER, BEEHIVE, EXPERIMENTAL, STROKES, THEATRICAL". Strictly pick theme only from these, Don't expect user to give exact word, pick relavent one from the list. If user asks completely irrelavent theme among the listed, tell user that those are only supported themes.\n- If user asks renaming a slideshow, it indicates the update in slideshow title.\n- Strictly scope your questions to Zoho Analytics and Zoho Show\n- Strictly Only Add, Delete and all Update operations are supported. If user tries other operations/actions, reply question as "This Operation is currently not supported. Please try a different request or contact support for help."\n\nOutput Format:\njson\n{\n  "presentationInformation": {\n    // Include modified key-value pairs for presentation information here\n  },\n  "slidesInformation": [\n    // Include modified or added slide information here\n    // Use empty objects {} to indicate no changes for specific slides\n  ]\n}\n\n\n\nExample 1:\n\nInput Slideshow JSON Object:\n{"presentationInformation":{"title":"Sales Trend Presentation","theme":"WHITEPAPER","withlogin":true,"slidecount":10},"slidesInformation":[{"slideId":1,"title":"Weekly Sales","includeinsights":false,"viewId":[{"title":"Sales Pie","id":"100003000000013455","type":"report"}]},{"slideId":2,"title":"Sales Order","includeinsights":false,"viewId":[{"title":"Sales Order","id":"100003000000013303","type":"report"}]},{"slideId":3,"title":"Sales analysis","includeinsights":false,"viewId":[{"title":"Sales Dashboard","id":"100003000000014248","type":"report"}]},{"slideId":4,"title":"sales by status","includeinsights":false,"viewId":[{"title":"sales by status","id":"100003000000013577","type":"report"}]},{"slideId":5,"title":"status-wise sales","includeinsights":false,"viewId":[{"title":"status-wise sales","id":"100003000000013567","type":"report"}]},{"slideId":6,"title":"Sales Across Years Info","includeinsights":false,"viewId":[{"title":"Sales Across Years","id":"100003000000014113","type":"report"}]},{"slideId":7,"title":"sales Across Month","includeinsights":false,"viewId":[{"title":"sales Across Month","id":"100003000000014043","type":"report"}]},{"slideId":8,"title":"ship mode-wise Sales","includeinsights":false,"viewId":[{"title":"ship mode-wise Sales","id":"100003000000013479","type":"report"}]},{"slideId":9,"title":"Distribution of Sales","includeinsights":false,"viewId":[{"title":"Distribution of Sales","id":"100003000000014110","type":"report"}]},{"slideId":10,"title":"total Sales by region","includeinsights":false,"viewId":[{"title":"total Sales by region","id":"100003000000013425","type":"report"}]}]}\n\nBelow are different usecases for the Input of Example 1\n\nUsecase 1:\nUser Input 1: Add a slide using monthly sales views\nOutput: {"presentationInformation":{"slidecount":11},"slidesInformation":[{"slideId":11,"added":true,"viewType":"view","viewName":"monthly sales"}]}\n\nExplanation of usecase 1: Here, We are adding a new slide. This new slide will have profit view. So a new slide is added in slidesInformation and the slidecount is increased.\n\nUsecase 2:\nUser Input 1: Add a slide after 1st slide using sales by region views\nOutput: {"presentationInformation":{"slidecount":11},"slidesInformation":[{"slideId":2,"added":true,"viewType":"view","viewName":"sales by region"}]}\n\n\nUsecase 3:\nUser Input 1: Change the slideshow theme to white paper\nOutput: {"presentationInformation":{"theme":"WHITEPAPER"},"slidesInformation":[]}\n\n\n\nUsecase 4:\nUser Input 1: Add a slide at top using region reports\nOutput: {"presentationInformation":{"slidecount":11},"slidesInformation":[{"slideId":1,"added":true,"viewType":"report","viewName":"region"}]}\n\n\nUsecase 5:\nUser Input 1: delete last slide\nOutput: {"presentationInformation":{"slidecount":9},"slidesInformation":[{"slideId":10,"deleted":true}]}\n\nUsecase 6:\nUser Input 1: delete the last slide and add a slide using profit reports\nOutput: {"presentationInformation":{},"slidesInformation":[{"slideId":10,"deleted":true},{"slideId":10,"added":true,"viewType":"report","viewName":"profit"}]}\n\nUsecase 7:\nUser Input 1: delete first slide and add a slide using profit reports\nOutput: {"presentationInformation":{},"slidesInformation":[{"slideId":1,"deleted":true},{"slideId":10,"added":true,"viewType":"report","viewName":"profit"}]}\n\n\nUsecase 8:\nUser Input 1: delete\nOutput: {"question":"What you want to delete?"}\n\n\nUsecase 9:\nUser Input 1: delete a slide\nOutput: {"question":"Which slide you want to delete?"}\nUser Input 2: last one\nOutput: {"presentationInformation":{"slidecount":9},"slidesInformation":[{"slideId":10,"deleted":true}]}\n\n\nUsecase 10:\nUser Input 1: update the theme\nOutput: {"question":"These are the supported themes: WHITEPAPER, BEEHIVE, EXPERIMENTAL, STROKES, THEATRICAL. Pick one from these"}\nUser Input 2: experiment\nOutput: {"presentationInformation":{"theme":"EXPERIMENTAL"},"slidesInformation":[]}\n\n\nUsecase 11:\nUser Input 1: delete 5th slide\nOutput: {"presentationInformation":{"slidecount":9},"slidesInformation":[{"slideId":5,"deleted":true}]}\n\n\nUsecase 12:\nUser Input 1: add sales report as new slide\nOutput: {"presentationInformation":{"slidecount":11},"slidesInformation":[{"slideId":11,"added":true,"viewType":"report","viewName":"sales"}]}\n\nUsecase 13:\nUser Input 1: Change the slideshow theme to light\nOutput: {"question":"These are the supported themes: WHITEPAPER, BEEHIVE, EXPERIMENTAL, STROKES, THEATRICAL. Pick one from these"}\n\nUsecase 14:\nUser Input 1: update the theme\nOutput: {"question":"These are the supported themes: WHITEPAPER, BEEHIVE, EXPERIMENTAL, STROKES, THEATRICAL. Pick one from these"}\nUser Input 2: black\nOutput: {"question":"These are the supported themes: WHITEPAPER, BEEHIVE, EXPERIMENTAL, STROKES, THEATRICAL. Pick one from these"}\n\n\nUsecase 15:\nUser Input 1: insert charts of sales as a slide\nOutput: {"presentationInformation":{"slidecount":11},"slidesInformation":[{"slideId":11,"added":true,"viewType":"report","viewName":"sales"}]}\n\n\nUsecase 16:\nUser Input 1: Add analysis view as a slide at 2nd index\nOutput: {"presentationInformation":{"slidecount":11},"slidesInformation":[{"slideId":2,"added":true,"viewType":"view","viewName":"analysis"}]}\n\n\nUsecase 17:\nUser Input 1: Add sales view slide at 4th position as a slide\nOutput: {"presentationInformation":{"slidecount":11},"slidesInformation":[{"slideId":4,"added":true,"viewType":"view","viewName":"sales"}]}\n\n\nUsecase 18:\nUser Input 1: Add sales views\nOutput: {"presentationInformation":{"slidecount":11},"slidesInformation":[{"slideId":11,"added":true,"viewType":"view","viewName":"sales"}]}\n\n\nUsecase 19:\nUser Input 1: Add sales views to the slideshow\nOutput: {"presentationInformation":{"slidecount":11},"slidesInformation":[{"slideId":11,"added":true,"viewType":"view","viewName":"sales"}]}\n\n\nUsecase 20:\nUser Input 1: Add sales report to the slideshow\nOutput: {"presentationInformation":{"slidecount":11},"slidesInformation":[{"slideId":11,"added":true,"viewType":"report","viewName":"sales"}]}\n\n\nUsecase 21:\nUser Input 1: rename the slideshow to Sales analysis\nOutput: {"presentationInformation":{"title":"Sales analysis"},"slidesInformation":[]}\n\n\nUsecase 22:\nUser Input 1: rename the 1st slide to Sales analysis\nOutput: {"question":"Renaming a slide is currently not supported."}\n\nUsecase 22:\nUser Input 1: get me link of this slideshow\nOutput: {"presentationInformation":{"withlogin":"true"},"slidesInformation":[]}\n\n\nUsecase 23:\nUser Input 1: delete this slideshow/delete it/delete this\nOutput: {"presentationInformation":{"deleted":true},"slidesInformation":[]}\n\n\nExample 2:\n\nInput Slideshow JSON Object:\n{"presentationInformation":{"title":"Sales Trend Presentation","theme":"WHITEPAPER","withlogin":true,"slidecount":3},"slidesInformation":[{"slideId":1,"title":"Weekly Sales","includeinsights":false,"viewId":[{"title":"status-wise loss","id":"100003000000013575","type":"report"}]},{"slideId":2,"title":"Weekly Sales","includeinsights":false,"viewId":[{"title":"sales profit loss by year","id":"100003000000013448","type":"report"}]},{"slideId":3,"title":"Weekly Sales","includeinsights":false,"viewId":[{"title":"total profit and total loss","id":"100003000000013342","type":"report"}]}]}\n\nThis Input json will have three slides.\n\n\nUsecase 1:\nUser Input 1: delete one slide\nOutput: {"question":"Which slide you want to delete?"}\nUser Input 2: first one\nOutput: {"presentationInformation":{"slidecount":2},"slidesInformation":[{"slideId":1,"deleted":true}]}\n\n\n\nUsecase 2:\nUser Input 1: delete 2nd slide\nOutput: {"presentationInformation":{"slidecount":2},"slidesInformation":[{"slideId":2,"deleted":true}]}\n\n\n\nUsecase 3:\nUser Input 1: delete middle slide\nOutput: {"presentationInformation":{"slidecount":2},"slidesInformation":[{"slideId":2,"deleted":true}]}\n\n\nUsecase 4:\nUser Input 1: Add a slide using profit views\nOutput: {"presentationInformation":{"slidecount":4},"slidesInformation":[{"slideId":4,"added":true,"viewType":"view","viewName":"profit"}]}\n\n\n\n\nExample 3:\n\nInput Slideshow JSON Object:\n{"presentationInformation":{"title":"Sales Trend Presentation","theme":"WHITEPAPER","withlogin":true,"slidecount":18},"slidesInformation":[{"slideId":1,"title":"Sales Pie","includeinsights":false,"viewId":[{"title":"Sales Pie","id":"100003000000013455","type":"report"}]},{"slideId":2,"title":"Sales Order","includeinsights":false,"viewId":[{"title":"Sales Order","id":"100003000000013303","type":"table"}]},{"slideId":3,"title":"Sales","includeinsights":false,"viewId":[{"title":"Sales","id":"100003000000014248","type":"dashboard"}]},{"slideId":4,"title":"sales by status","includeinsights":false,"viewId":[{"title":"sales by status","id":"100003000000013577","type":"report"}]},{"slideId":5,"title":"status-wise sales","includeinsights":false,"viewId":[{"title":"status-wise sales","id":"100003000000013567","type":"report"}]},{"slideId":6,"title":"Sales Across Years","includeinsights":false,"viewId":[{"title":"Sales Across Years","id":"100003000000014113","type":"report"}]},{"slideId":7,"title":"sales Across months","includeinsights":false,"viewId":[{"title":"sales Across months","id":"100003000000014043","type":"report"}]},{"slideId":8,"title":"ship mode-wise Sales","includeinsights":false,"viewId":[{"title":"ship mode-wise Sales","id":"100003000000013479","type":"report"}]},{"slideId":9,"title":"Distribution of Sales","includeinsights":false,"viewId":[{"title":"Distribution of Sales","id":"100003000000014110","type":"report"}]},{"slideId":10,"title":"total Sales by region","includeinsights":false,"viewId":[{"title":"total Sales by region","id":"100003000000013425","type":"report"}]},{"slideId":11,"title":"total Sales by year","includeinsights":false,"viewId":[{"title":"total Sales by year","id":"100003000000013425","type":"report"}]},{"slideId":12},{"slideId":13},{"slideId":14},{"slideId":15},{"slideId":16},{"slideId":17},{"slideId":18}]}\n\n\nUsecase 1:\nUser Input 1: Add analysis view as a slide at 2nd index\nOutput: {"presentationInformation":{"slidecount":19},"slidesInformation":[{"slideId":2,"added":true,"viewType":"view","viewName":"analysis"}]}\n\n\nUsecase 2:\nUser Input 1: Add sales view slide at 4th position as a slide\nOutput: {"presentationInformation":{"slidecount":19},"slidesInformation":[{"slideId":4,"added":true,"viewType":"view","viewName":"sales"}]}\n\nUsecase 3:\nUser Input 1: Add a slide using monthly sales views\nOutput: {"presentationInformation":{"slidecount":19},"slidesInformation":[{"slideId":19,"added":true,"viewType":"view","viewName":"monthly sales"}]}\n\n\nUsecase 4:\nUser Input 1: delete 5th slide\nOutput: {"presentationInformation":{"slidecount":17},"slidesInformation":[{"slideId":5,"deleted":true}]}\n\n\nUsecase 5:\nUser Input 1: delete last slide\nOutput: {"presentationInformation":{"slidecount":17},"slidesInformation":[{"slideId":18,"deleted":true}]}\n\n\n\n\n\nExample cases where questions are not needed for view name:\nIf the sentence has this pattern like <view-name> view/views/report/chart\n1. Add a slide using monthly sales views\n2. Add a slide after 1st slide using sales by region views\n3. Add a slide using sales view\n4. Add a slide for profit views\n5. hey can you add a profit view as slide\n\n\nUse only the below questions:\n1. Which slide you want to delete?\n2. Provide me the view details you want to add in the new slide\n3. These are the supported themes: WHITEPAPER, BEEHIVE, EXPERIMENTAL, STROKES, THEATRICAL. Pick one from these\n4. Provide the title of the slideshow
za.askzia.criteriameta.prompt.for.zia.turbo=You are given a JSON response from an LLM containing sections such as filter, group_by, and aggregate, along with other relevant metadata. One of the filters includes a column that appears incorrectly mapped, as its values do not align well with the provided filter value. However, a list of alternative columns that directly match the filter value is available. You are also provided with sample data and criteria from the originally mapped column.\r\nYour task is to:\r\n1. Evaluate whether the originally mapped column aligns with the filter value and its associated data. If it does, retain it and add the applicable criteria value under a new key named \"criteria\" in the filter.\r\n2. If the mapped column does not align well with the filter value, select the most appropriate column from the list of alternatives.\r\n3. Ensure the selected column is also compatible with the context of the aggregate and group_by sections in the JSON.\r\n4. Return the final selected column name.\r\n\r\nMake sure the column you return is the best semantic and structural fit for the filter value based on the overall JSON context and sample data provided.\r\n\r\nInput JSON Structure:\r\n{\r\n \"filter\": [\r\n  {\r\n \"function\": \"(function | datefunction | null)\",\r\n \"column\": null,\r\n \"table\": null,\r\n \"type\": null,\r\n \"value\": null,\r\n \"excludeFilter\": null\r\n  }\r\n ],\r\n \"group_by\": [\r\n  {\r\n \"function\": null,\r\n \"datefunction\": null,\r\n \"sortType\": null,\r\n \"column\": null,\r\n \"axisposition\": null,\r\n \"table\": null\r\n  }\r\n ],\r\n \"aggregate\": [\r\n  {\r\n \"function\": [],\r\n \"column\": null,\r\n \"axisposition\": null,\r\n \"table\": null,\r\n \"sortType\": null,\r\n \"windowfunction\": []\r\n  }\r\n ],\r\n \"whType\": null,\r\n \"viewType\": null,\r\n \"concept\": null,\r\n \"isComplexQuery\": true | false\r\n}\r\n\r\nMatched Columns Format:\r\n[\r\n {\"clusterID\": 1, \"tableName\": \"Table1\", \"columnName\": \"Column1\"},\r\n {\"clusterID\": 2, \"tableName\": \"Table2\", \"columnName\": \"Column2\"}\r\n]\r\n\r\nLLM Response Column Matches and Criteria Format:\r\n{\r\n \"columns\": [\r\n  {\r\n \"columnID\": \"ColumnID1\",\r\n \"tableID\": \"TableID1\",\r\n \"criteria\": [\"Value1\", \"Value2\", \"Value3\"]\r\n  },\r\n  {\r\n \"columnID\": \"ColumnID2\",\r\n \"tableID\": \"TableID2\",\r\n \"criteria\": [\"Value4\", \"Value5\"]\r\n  }\r\n ]\r\n}\r\n\r\nYou are given a JSON response from a language model that includes sections such as filter, group_by, and aggregate, along with metadata. Your task is to validate and possibly correct the filter column based on the filter value, sample data, and the overall context.\r\n\u2028Steps to Follow:\r\n1. Analyze the filter, group_by, and aggregate sections of the input JSON.\r\n2. Evaluate each candidate column (including the initially mapped column and other matched columns) for its compatibility with the filter value and its relevance to the group_by and aggregate sections.\r\n3. If the originally mapped column contains data that aligns with the filter value and associated criteria, retain this column.\r\n    * In that case, include only the applicable single value in \"data\" key under the \"criteria\" key in the output JSON.\r\n4. If a better matching column exists (based on semantic similarity, filter value match, and compatibility with the aggregation or grouping context), replace the originally mapped column with this new one.Strictly pick the filter which as similar meaning to the criteria user have asked from the criteria list. Check all the criteria and provide back with best similar and matching to the user asked criteria.\r\n    * Ensure the selected criteria value is the closest possible match to the one previously chosen.\r\n5. For negation queries (e.g., containing words like not, exclude), if the value is not present in the actual data, set \"excludeFilter\": true; otherwise, use false.\r\n6. Return only the final selected column name and the corresponding criteria value (if applicable) in the following JSON format:\r\n\r\nOutput JSON structure:\r\n{\r\n \"clusterID\": null,\r\n \"tableName\": null,\r\n \"columnName\": null,\r\n \"criteria\":{ \"data\":null, \"excludeFilter\":true\/false}\r\n}\r\nStrictly give the output in the above format and strictly criteria must be in json object .\r\nReturn only one selected column and one best-matching string value in criteria.data.\r\nThe criteria.data must be a string, not a JSON object or array.\r\nNo explanations or additional text , only the output in the exact JSON format above.\r\n
za.askzia.complex_query_extraction.prompt.for.zia.turbo=Objective: Given a conversation history containing queries, their associated metadata (such as measures, functions, tables, and filters), and the generated responses\u2014for Example : [{\"role\":\"user\",\"content\":\"What is the correlation between sales and shipping cost?\"},{\"role\":\"assistant\",\"content\":{\"meta\":{\"measures\":[{\"cluster_id\":1,\"table_name\":\"testdata\",\"column_name\":\"Sales\",\"function\":\"sum\"},{\"cluster_id\":1,\"table_name\":\"testdata\",\"column_name\":\"Shipping Cost\",\"function\":\"sum\"}]},\"response\":{\"string_response\":\"The correlation between sales and shipping cost is 0.75.\"}}},{\"role\":\"user\",\"content\":\"what is the correlation of the above measures in East regions?\"}]\u2014extract the structured metadata and identify the analysis method for the latest query by leveraging context from previous interactions. Fill in any missing details in the latest query using earlier metadata and responses, while preserving previously applied criteria unless explicitly changed by the user. Generate a refined version of the latest query that is complete, contextually accurate, and unambiguous for further processing. Generate a valid SQL query to retrieve the required data, and, if necessary, produce Python code or apply a static algorithm for deeper analysis. Processing Logic: Determine Query Complexity - If the query can be answered using only SQL (e.g., simple aggregation, filtering), return \"sql_query\" and set \"python_code\": []. If deeper analysis is required (e.g., correlation, forecasting, impact), check if a static analysis method is applicable. Use Static Algorithm If Applicable (Preferred) - If the data meets input requirements: set \"analysis_method\", \"used_data_frame\", \"input_columns\", and optional configurations for each analysis methods. Do not generate \"python_code\" if static method is used. If Static Method Not Applicable \u2192 Generate Dynamic Python - Use SQLite-compatible SQL and generate a single formatted string inside \"python_code\". Best Practices: use pd.to_datetime(..., errors='coerce'), pd.to_numeric(..., errors='coerce'), drop NaNs, and validate inputs. SQL Generation Rules: Use strftime('%Y-%m', \"date_column\") for monthly grouping, strftime('%Y', \"date_column\") for yearly grouping, and use double quotes \" \" for column names with spaces.\r\nInstructions for Query Refinement\r\nWhile refining the user query based on the conversation context, do consider only the meta and not report names and related stuff.\r\nSupported Static Analysis Methods (with Examples): \r\n1. Correlation - Mandatory: 2 Measures, Configuration -  \"correlation_configuration\": {\"correlation_algorithm\":\"PEARSONS (default) or KENDALLS or SPEARMANS\"}, Example Queries: \"What is the correlation between Sales and Profit since 2020?\", \"Does an increase in Sales lead to an increase in Profit?\"\r\n2. Category Correlation - Mandatory: 2 Measures, 1 Dimension (Date or Non-Date), Configuration - \"correlation_configuration\": {\"periodicity\":\"yearly or quarterly or monthly or weekly or daily or <null>\",\"correlation_algorithm\":\"PEARSONS (default) or KENDALLS or SPEARMANS\"}, Example Queries: \"What is the correlation between Sales and Profit in each Region?\", \"How strongly are Sales and Profit correlated in East Region?\", \"What is the monthly correlation between sales and profit?\r\n3. Concentration - Mandatory: 1 Measure, either 1 Non-Date Dimension or 1 Date Dimension.  Example Queries: \"What is the contribution of each Region to total sales?\", \"Which Customer Type accounts for 20% of profit?\", \"List top 5 products by sales contribution\", \"List the offsetting products in sales\", \"show the month-wise contributions to sales\", \"List the bottom 5 products in terms of sales\", \"yearly performance of products\/regions\"\r\n\"What is concentration?\"\r\nConcentration refers to the extent to which a specific measure (such as sales, profit, or revenue) is distributed or aggregated across the values of a single dimension\u2014either a non-date dimension (like Product, Region, or Customer Type) or a date dimension (like Month or Year). It helps identify which dimension members contribute most or least to the total value of the measure.\r\nThis also includes analyzing yearly or monthly performance\u2014such as how different products perform in terms of sales across different years\u2014making temporal concentration a key part of this analysis.\r\n4. Outlier - Mandatory: 1 Measure, 1 Date Dimension. Example Queries: \"Identify outliers in sales\", \"What is the average deviation in sales outliers for 2015?\" \r\n5. Category Outlier - Mandatory: 1 Measure, 1 Date Dimension, 1 Non-Date Dimension, Example Queries: \"Is there any outliers in Finland's sales figures during 2022 to 2024?\", \"list the anomalies in sales for Canada and Greenland\"\r\n6. Forecast - Mandatory: 1 Measure, 1 Date Dimension, Configuration - \"forecast_configuration\" : {\"forecast_period\":3}. Fill \"forecast_period\" with the number of points to be forecasted and the default value is 3.\r\nExample Queries: \"What is the forecast of total sales for the next 3 years?\", \"Predict sales for the next 3 quarters\"\r\n7. Category Forecast - Mandatory: 1 Measure, 1 Date Dimension, 1 Non-Date Dimension, Configuration - \"forecast_configuration\" : {\"forecast_period\":3}. Fill \"forecast_period\" with the number of points to be forecasted and the default value is 3. \r\nExample Queries: \"Forecast total sales for each region\", \"What are the estimated sales per region for the next 3 years?\"\r\n8. Trend - Mandatory: 1 Measure, 1 Date Dimension, Example Queries: \"What is the trend in total sales?\", \"Does sales show an uptrend over time?\"\r\n9. Category Trend - Mandatory: 1 Measure, 1 Date Dimension, 1 Non-Date Dimension, Example Queries: \"What is the trend of total sales by Region?\", \"How many regions show an upward trend in sales?\"\r\n10. Time Over Time - Mandatory: 1 Measure, 1 Date Dimension. Example Queries: \"Get Year-over-Year growth for sales\", \"What\u2019s the month-on-month sales performance?\"\r\n11. Category Time Over Time - Mandatory: 1 Measure, 1 Date Dimension, 1 Non-Date Dimension, Example Queries: \"What is the year-over-year growth in sales in Canada and Greenland?\"\r\n12. Skewness - Mandatory: 1 Measure, 1 Dimension (Date or Non-Date), Example Queries: \"How are the sales across regions skewed?\", \"Is the sales across regions right skewed or left skewed?\"\r\n13. Seasonality - Mandatory: 1 Measure, 1 Date Dimension, Example Queries: \"What is the seasonal pattern shown by monthly sales?\", \"Does monthly sales show any seasonal patterns?\"\r\n14. Category Seasonality - Mandatory: 1 Measure, 1 Date Dimension, 1 Non-Date Dimension, Example Queries: \"What is the seasonal pattern shown by monthly sales across regions?\", \"Does quarterly sales across regions show any seasonal patterns?\"\r\n15. Diagnosis - Mandatory Inputs: 1 Measure, 1 Date Dimension, Target Date, Comparator Date (optional Target Influencer). Comparator date can be determined in two ways: it may either be explicitly specified in the query or, by default, set as the date immediately preceding the target date. Both target date and comparator date are represented as JSON arrays, and each comparator date should be mapped to the corresponding index of the target date array. Pick the target influencers from the conversation context if not explicitly mentioned in the query.\r\nConfiguration - diagnosis_configuration:\r\n{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"<string> - Name of the table (e.g., 'Sales Order')\",\"column_name\":\"<string> - Name of the date column (e.g., 'order-date')\",\"sub_type\":\"individual\",\"date_type\":\"<string> - One of [year or absquarter or absmonth or absweek or date or datetime']\",\"values\":[\"2022 or Q1 2022 or Jan 2022 or W1 2022 or 1 Jan 2022 or 1 Jan 2022 14:00:00\"]}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"<string> - Name of the table (e.g., 'Sales Order')\",\"column_name\":\"<string> - Name of the date column (e.g., 'order-date')\",\"sub_type\":\"individual\",\"date_type\":\"<string> - One of [year or absquarter or absmonth or absweek or date or datetime']\",\"values\":[\"2022 or Q1 2022 or Jan 2022 or W1 2022 or 1 Jan 2022 or 1 Jan 2022 14:00:00\"]}],\"target_influencer\":[{\"type\":\"string_criteria\",\"table_name\":\"<table name>\",\"column_name\":\"<column name>\",\"values\":[\"...\"]}],\"direction\":\"positive\/negative (based on the intent of the user)\"}\r\nExample: \r\n#1 \"Why did sales decreased in 2023?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[],\"direction\":\"negative\",\"type\":\"one_vs_one\"} \r\n#2 \"Why did sales increased in Finland during 2023?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[],\"direction\":\"positive\",\"type\":\"one_vs_one\"} \r\n#3 \"Find the key influencers for the increase in sales in 2023 with help of Country, Customers and Shipping Mode\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[{\"function\":\"actual\",\"table_name\":\"Sales\",\"column_name\":\"Country\"},{\"function\":\"actual\",\"table_name\":\"Sales\",\"column_name\":\"Customer\"},{\"function\":\"actual\",\"table_name\":\"Sales\",\"column_name\":\"Shipping Mode\"}],\"direction\":\"positive\",\"type\":\"one_vs_one\"} \r\n#4 \"Why 2023 sales is an anomaly?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":false}],\"explain_by_columns\":[],\"direction\":\"negative\",\"type\":\"one_vs_others\"} \r\n#5 \"How Finland impacted the drop in sales in 2023?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[],\"target_influencer\":[{\"type\":\"string_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Country\",\"values\":[\"Finland\"],\"include\":true}],\"direction\":\"negative\",\"type\":\"one_vs_one\"} \r\n#6 \"How Country impacted the drop in sales in 2023?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[],\"target_influencer\":[{\"type\":\"string_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Country\",\"values\":[],\"include\":true}],\"direction\":\"negative\",\"type\":\"one_vs_one\"}\r\n#7 \"How 20 to 30 range from Discount impacted the drop in sales in 2023?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[],\"target_influencer\":[{\"type\":\"string_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Discount\",\"values\":[\"20 to 30\"],\"include\":true}],\"direction\":\"negative\",\"type\":\"one_vs_one\"} \r\n#8 \"How Online shopping mode impacted the increase in sales in 2024?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2024\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"explain_by_columns\":[],\"target_influencer\":[{\"type\":\"string_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Shopping Mode\",\"values\":[\"Online\"],\"include\":true}],\"direction\":\"positive\",\"type\":\"one_vs_one\"} \r\n#9 \"How all the drivers impacted the increase in 2023?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[],\"target_influencer\":[\"<influencers from conversation context>\"],\"direction\":\"negative\",\"type\":\"one_vs_one\"}\r\n16. Recommendation - Mandatory Inputs: 1 Measure\r\nExamples: How to improve sales\r\n17. Insights - Mandatory Inputs: At least one measure column. Configuration - \"insights_config\":{\"verbosity\":\"low\/medium\/high\",\"language\":\"english\/french\/spanish\",\"view_type\":\"descriptive\/visual\",\"focussed_meta\":{\"focussed_measure\":{\"function\":\"<measure function>\",\"table_name\":\"<name of the table>\",\"column_name\":\"<name of the column>\"},\"focussed_criteria\":[]}}\r\nExamples: \r\n#1 \"Get me insights for sales across products in Finland\" - {\"insights_config\":{\"verbosity\":\"medium\",\"language\":\"english\",\"view_type\":\"descriptive\",\"focussed_meta\":null}} \r\n#2 \"Get me insights for sales across products in 2023 and 2024\" - {\"insights_config\":{\"verbosity\":\"medium\",\"language\":\"english\",\"view_type\":\"descriptive\",\"focussed_meta\":null}} \r\n#3 \"Get me insights for the above report\" - {\"insights_config\":{\"verbosity\":\"medium\",\"language\":\"english\",\"view_type\":\"descriptive\",\"focussed_meta\":null}} \r\nNote: The metadata for this query should be derived from the opened or selected report metadata provided in the conversation history.\r\n#4 \"narrate the insights in spanish\" - \r\n{\"insights_config\":{\"verbosity\":\"medium\",\"language\":\"spanish\",\"view_type\":\"descriptive\",\"focussed_meta\":null}} \r\n#5 \"Get more insights for sales across products\" - \r\n{\"insights_config\":{\"verbosity\":\"high\",\"language\":\"english\",\"view_type\":\"descriptive\",\"focussed_meta\":null}} \r\n#6 \"Get deeper insights for sales across products\" - \r\n{\"insights_config\":{\"verbosity\":\"high\",\"language\":\"english\",\"view_type\":\"descriptive\",\"focussed_meta\":null}} \r\n#7 \"Get verbose insights for sales across products in 2023\" - \r\n{\"insights_config\":{\"verbosity\":\"high\",\"language\":\"english\",\"view_type\":\"descriptive\",\"focussed_meta\":null}} \r\n#8 \"Explain Finland's sales\" - {\"insights_config\":{\"verbosity\":\"medium\",\"language\":\"english\",\"view_type\":\"descriptive\",\"focussed_meta\":{\"focussed_measure\":{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Sales\"},\"focussed_criteria\":[{\"table_name\":\"testdata\",\"column_name\":\"Country\",\"type\":\"string_criteria\",\"values\":[\"Finland\"]}]}}}\r\n#9 \"Explain Finland's sales in 2023\" - {\"insights_config\":{\"verbosity\":\"medium\",\"language\":\"english\",\"view_type\":\"descriptive\",\"focussed_meta\":{\"focussed_measure\":{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Sales\"},\"focussed_criteria\":[{\"table_name\":\"testdata\",\"column_name\":\"Country\",\"type\":\"string_criteria\",\"values\":[\"Finland\"]},{\"table_name\":\"testdata\",\"column_name\":\"Date\",\"type\":\"date_criteria\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"]}]}}}\r\n#10 \"Explain 2023 sales\" - {\"insights_config\":{\"verbosity\":\"medium\",\"language\":\"english\",\"view_type\":\"descriptive\",\"focussed_meta\":{\"focussed_measure\":{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Sales\"},\"focussed_criteria\":[{\"table_name\":\"testdata\",\"column_name\":\"Date\",\"type\":\"date_criteria\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"]}]}}}\r\n#11 \"Get me top insights for sales across products\" - {\"insights_config\":{\"verbosity\":\"medium\",\"language\":\"english\",\"view_type\":\"descriptive\",\"focussed_meta\":null}} \r\n\r\n18. Inter-Category Correlation - Mandatory: 1 Measure, 1 Non-date dimension, Configuration - \"correlation_configuration\": {\"correlation_algorithm\":\"PEARSONS (default) or KENDALLS or SPEARMANS\"}, Example Queries: \"How East and West Region's sales correlated?\", \"Is there any relation in changes of East and West Region's sales?\"\r\nNote: All 'why' queries fall under the 'diagnosis' analysis method.\r\nFiltering Schema (with Examples)\r\n1. Individual Date Criteria - \r\n{\"table_name\":\"testdata\",\"column_name\":\"date column\",\"type\":\"date_criteria\",\"sub_type\":\"individual\",\"date_type\":\"year or absquarter or absmonth or absweek or date or datetime\",\"values\":[\"2022 or Q1 2022 or Jan 2022 or W1 2022 or 1 Jan 2022 or 1 Jan 2022 14:00:00\"],\"include\":\"<true> or <false>\"}, \r\nExample Queries: \r\n#1 What is the correlation between sales and ad spend in August 2022?\r\n#2 What is the correlation between sales and ad spend in the 2nd week of August 2022? - {\"table_name\":\"testdata\",\"column_name\":\"Order Date\",\"type\":\"date_criteria\",\"sub_type\":\"individual\",\"date_type\":\"absweek\",\"values\":[\"W32 2022\"],\"include\":\"<true> or <false>\"}\r\n#3 In 2022 how Ad spend related to Sales and then in 2023 How it was? - {\"table_name\":\"testdata\",\"column_name\":\"Order Date\",\"type\":\"date_criteria\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\",\"2023\"],\"include\":\"<true> or <false>\"}\r\nNote: If the measures and dimensions are of the same table, then include all the criteria under one single bucket unless meta changes\r\n2. Range Date Criteria - {\"table_name\":\"testdata\",\"column_name\":\"date column\",\"type\":\"date_criteria\",\"sub_type\":\"range\",\"date_type\":\"day\",\"range\":{\"from\":{\"day\":1,\"month\":1,\"year\":2022},\"to\":{\"day\":31,\"month\":12,\"year\":2023}},\"include\":\"<true> or <false>\"} \r\nNote: Strictly follow the above range format for all date functions in Range Date Criteria\r\nExample Queries:\r\n#1 What is the correlation between sales and ad spend during Jan 2022 to Jul 2022?\r\n#2 What is the correlation between sales and ad spend from the 3rd to the 6th week of 2022?\r\n3. Seasonal Date Criteria - {\"table_name\":\"testdata\",\"column_name\":\"date column\",\"type\":\"date_criteria\",\"sub_type\":\"seasonal\",\"date_type\":\"quarter or month or week or weekday or day or hour\",\"values\":[\"Q1 or Aug or W2 or Sun or 27 or 14\"],\"include\":\"<true> or <false>\"}\r\nExample Queries:\r\n#1 What is the correlation between sales and ad spend in Augusts every year?\r\n#2 What is the correlation between sales and ad spend in the 2nd Week of each year?\r\n4. Relative Date Criteria - {\"table_name\":\"testdata\",\"column_name\":\"date column\",\"type\":\"date_criteria\",\"sub_type\":\"relative\",\"date_type\":\"year or quarter or month or week or day\",\"relative_position\":\"this or last or previous\",\"period\":\"n\",\"include\":\"<true> or <false>\"}\r\nExample Queries:\r\n#1 What is the correlation between sales and ad spend in last 5 months? - {\"table_name\":\"testdata\",\"column_name\":\"Date\",\"type\":\"date_criteria\",\"sub_type\":\"relative\",\"date_type\":\"month\",\"relative_position\":\"last\",\"period\":5,\"include\":true}\r\n#2 What is the correlation between sales and ad spend in previous 5 weeks? - {\"table_name\":\"testdata\",\"column_name\":\"Date\",\"type\":\"date_criteria\",\"sub_type\":\"relative\",\"date_type\":\"week\",\"relative_position\":\"previous\",\"period\":5,\"include\":true}\r\n#3 What is the correlation between sales and ad spend in this month? - {\"table_name\":\"testdata\",\"column_name\":\"Date\",\"type\":\"date_criteria\",\"sub_type\":\"relative\",\"date_type\":\"month\",\"relative_position\":\"this\",\"period\":1,\"include\":true} \r\n#4 What is the correlation between sales and ad spend in last month? - {\"table_name\":\"testdata\",\"column_name\":\"Date\",\"type\":\"date_criteria\",\"sub_type\":\"relative\",\"date_type\":\"month\",\"relative_position\":\"previous\",\"period\":1,\"include\":true}\r\n5. Text Criteria - {\"table_name\":\"testdata\",\"column_name\":\"text column\",\"type\":\"string_criteria\",\"values\":[\"...\"],\"include\":\"<true> or <false>\"}\r\nExample Queries:\r\n#1 What is the correlation between sales and ad spend in East region?\r\n#2 What is the correlation between sales and ad spend for the Grocery category?\r\nPossible Functions\r\nMeasure - \"sum\" (default), \"average\", \"count\", \"distinct count\", \"minimum\", \"maximum\", \"standard deviation\", \"variance\", \"mode\", \"median\" & \"percentile\", \r\nDate Dimension - \"year\", \"absquarter\", \"absmonth\" (default), \"absweek\", \"fulldate\", \r\nText Dimension - \"actual\"\r\nOutput Format\r\nAssign sort_type 'asc' or 'desc' if the particular sort type is explicitly mentioned else go with 'default'. Use sort_type 'asc' for keywords like 'off-setting', 'Drivers of Decline', 'Under-performers' and similar terms. For limit, assign the number of rows the user might expect based on the query; else, null. For example, if the query is based on months, then the limit should be 12 since we need to show all the months in the output.\r\nOutput JSON Structure: {\"user_query\":\"<refined user query based on conversational history>\",\"combinations\":[{\"rank\":1,\"data_extraction\":[{\"name\":\"some_extraction\",\"measure\":[{\"function\":\"<measure function>\",\"table_name\":\"name of the table\",\"column_name\":\"name of the column\"}],\"dimension\":[{\"function\":\"dimension function\",\"table_name\":\"name of the table\",\"column_name\":\"name of the column\"}],\"criteria\":[],\"sql_query\":\"sql query for data extraction\"}],\"analysis_method\":\"can be null if python code mode is picked else one of the analysis method mentioned\",\"used_data_frame\":\"sales_data\",\"input_columns\":{\"measure\":[{\"function\":\"<measure function>\",\"table_name\":\"name of the table\",\"column_name\":\"name of the column\"}],\"dimension\":[{\"function\":\"dimension function\",\"table_name\":\"name of the table\",\"column_name\":\"name of the column\"}]},\"output_format\":{\"sort_type\":\"asc or desc or default\",\"limit\":\"<null> or the number of rows the user might expect\"},\"<optional_config>\":{}}]}, with maximum of 5 combinations tried in ranked order. \r\nSample Output for Query \"What is the correlation between sales and shipping cost in 2023?\": \r\n{\"user_query\":\"What is the correlation between sales and shipping cost in 2023?\",\"combinations\":[{\"rank\":1,\"data_extraction\":[{\"name\":\"df_sales_shipping\",\"measure\":[{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Net Sales\"},{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Shipping Cost\"}],\"dimension\":[{\"function\":\"year\",\"table_name\":\"testdata\",\"column_name\":\"Order Date\"}],\"criteria\":[{\"table_name\":\"testdata\",\"column_name\":\"Order Date\",\"type\":\"date_criteria\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"sql_query\":\"SELECT strftime('%Y-%m', \\\"Order Date\\\") AS month, SUM(\\\"Net Sales\\\") AS total_sales, SUM(\\\"Shipping Cost\\\") AS total_shipping_cost FROM testdata WHERE strftime('%Y', \\\"Order Date\\\") = '2023' GROUP BY month\"}],\"analysis_method\":\"Correlation\",\"used_data_frame\":\"df_sales_shipping\",\"input_columns\":{\"measures\":[{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Net Sales\"},{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Shipping Cost\"}],\"dimensions\":[{\"function\":\"year\",\"table_name\":\"testdata\",\"column_name\":\"Order Date\"}]},\"output_format\":{\"sort_type\":\"default\",\"limit\":null},\"correlation_configuration\":{\"correlation_algorithm\":\"PEARSONS\"}},{\"rank\":2,\"data_extraction\":[{\"name\":\"df_sales_shipping\",\"measure\":[{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Sales Amount\"},{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Shipping Cost\"}],\"dimension\":[{\"function\":\"year\",\"table_name\":\"testdata\",\"column_name\":\"Order Date\"}],\"criteria\":[{\"table_name\":\"testdata\",\"column_name\":\"Order Date\",\"type\":\"date_criteria\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"sql_query\":\"SELECT strftime('%Y-%m', \\\"Order Date\\\") AS month, SUM(\\\"Sales Amount\\\") AS total_sales, SUM(\\\"Shipping Cost\\\") AS total_shipping_cost FROM testdata WHERE strftime('%Y', \\\"Order Date\\\") = '2023' GROUP BY month\"}],\"analysis_method\":\"Correlation\",\"used_data_frame\":\"df_sales_shipping\",\"input_columns\":{\"measures\":[{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Sales Amount\"},{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Shipping Cost\"}],\"dimensions\":[{\"function\":\"year\",\"table_name\":\"testdata\",\"column_name\":\"Order Date\"}]},\"output_format\":{\"sort_type\":\"default\",\"limit\":null},\"correlation_configuration\":{\"correlation_algorithm\":\"PEARSONS\"}}]}\r\nOutput - DOs\r\nStrictly follow the given output json structure\r\n\r\nOutput \u2013 DON'Ts\r\n\r\nDo not include any descriptions or processing steps in the output.\r\nAvoid adding responses in the output.

za.askzia.complex_query_extraction.qwen.prompt.for.zia.turbo=Objective: Given a conversation history containing queries, their associated metadata (such as measures, functions, tables, and filters), and the generated responses, extract the structured metadata and identify the analysis method for the latest query by leveraging context from previous interactions. Fill in any missing details in the latest query using earlier metadata and responses, while preserving previously applied criteria unless explicitly changed by the user. Generate a refined version of the latest query that is complete, contextually accurate, and unambiguous for further processing. Generate a valid SQL query to retrieve the required data, and, if necessary, produce Python code or apply a static algorithm for deeper analysis. Processing Logic: Determine Query Complexity - If the query can be answered using only SQL (e.g., simple aggregation, filtering), return \"sql_query\" and set \"python_code\": []. If deeper analysis is required (e.g., correlation, forecasting, impact), check if a static analysis method is applicable. Use Static Algorithm If Applicable (Preferred) - If the data meets input requirements: set \"analysis_method\", \"used_data_frame\", \"input_columns\", and optional configurations for each analysis methods. Do not generate \"python_code\" if static method is used. If Static Method Not Applicable \u2192 Generate Dynamic Python - Use SQLite-compatible SQL and generate a single formatted string inside \"python_code\". Best Practices: use pd.to_datetime(..., errors='coerce'), pd.to_numeric(..., errors='coerce'), drop NaNs, and validate inputs. SQL Generation Rules: Use strftime('%Y-%m', \"date_column\") for monthly grouping, strftime('%Y', \"date_column\") for yearly grouping, and use double quotes \" \" for column names with spaces.\r\nInstructions for Query Refinement\r\nWhile refining the user query based on the conversation context, do consider only the meta and not report names and related stuff.\r\nSupported Static Analysis Methods (with Examples): \r\n1. Correlation - Mandatory: 2 Measures, Configuration -  \"correlation_configuration\": {\"correlation_algorithm\":\"PEARSONS (default) or KENDALLS or SPEARMANS\"}, Example Queries: \"What is the correlation between Sales and Profit since 2020?\", \"Does an increase in Sales lead to an increase in Profit?\"\r\n2. Category Correlation - Mandatory: 2 Measures, 1 Dimension (Date or Non-Date), Configuration - \"correlation_configuration\": {\"periodicity\":\"yearly or quarterly or monthly or weekly or daily or <null>\",\"correlation_algorithm\":\"PEARSONS (default) or KENDALLS or SPEARMANS\"}, Example Queries: \"What is the correlation between Sales and Profit in each Region?\", \"How strongly are Sales and Profit correlated in East Region?\", \"What is the monthly correlation between sales and profit?\r\n3. Concentration - Mandatory: 1 Measure, either 1 Non-Date Dimension or 1 Date Dimension.  Example Queries: \"What is the contribution of each Region to total sales?\", \"Which Customer Type accounts for 20% of profit?\", \"List top 5 products by sales contribution\", \"List the offsetting products in sales\", \"show the month-wise contributions to sales\", \"List the bottom 5 products in terms of sales\", \"yearly performance of products\/regions\"\r\n\"What is concentration?\"\r\nConcentration refers to the extent to which a specific measure (such as sales, profit, or revenue) is distributed or aggregated across the values of a single dimension\u2014either a non-date dimension (like Product, Region, or Customer Type) or a date dimension (like Month or Year). It helps identify which dimension members contribute most or least to the total value of the measure.\r\nThis also includes analyzing yearly or monthly performance\u2014such as how different products perform in terms of sales across different years\u2014making temporal concentration a key part of this analysis.\r\n4. Outlier - Mandatory: 1 Measure, 1 Date Dimension. Example Queries: \"Identify outliers in sales\", \"What is the average deviation in sales outliers for 2015?\" \r\n5. Category Outlier - Mandatory: 1 Measure, 1 Date Dimension, 1 Non-Date Dimension, Example Queries: \"Is there any outliers in Finland's sales figures during 2022 to 2024?\", \"list the anomalies in sales for Canada and Greenland\"\r\n6. Forecast - Mandatory: 1 Measure, 1 Date Dimension, Configuration - \"forecast_configuration\" : {\"forecast_period\":3}. Fill \"forecast_period\" with the number of points to be forecasted and the default value is 3.\r\nExample Queries: \"What is the forecast of total sales for the next 3 years?\", \"Predict sales for the next 3 quarters\"\r\n7. Category Forecast - Mandatory: 1 Measure, 1 Date Dimension, 1 Non-Date Dimension, Configuration - \"forecast_configuration\" : {\"forecast_period\":3}. Fill \"forecast_period\" with the number of points to be forecasted and the default value is 3. \r\nExample Queries: \"Forecast total sales for each region\", \"What are the estimated sales per region for the next 3 years?\"\r\n8. Trend - Mandatory: 1 Measure, 1 Date Dimension, Example Queries: \"What is the trend in total sales?\", \"Does sales show an uptrend over time?\"\r\n9. Category Trend - Mandatory: 1 Measure, 1 Date Dimension, 1 Non-Date Dimension, Example Queries: \"What is the trend of total sales by Region?\", \"How many regions show an upward trend in sales?\"\r\n10. Time Over Time - Mandatory: 1 Measure, 1 Date Dimension. Example Queries: \"Get Year-over-Year growth for sales\", \"What\u2019s the month-on-month sales performance?\"\r\n11. Category Time Over Time - Mandatory: 1 Measure, 1 Date Dimension, 1 Non-Date Dimension, Example Queries: \"What is the year-over-year growth in sales in Canada and Greenland?\"\r\n12. Skewness - Mandatory: 1 Measure, 1 Dimension (Date or Non-Date), Example Queries: \"How are the sales across regions skewed?\", \"Is the sales across regions right skewed or left skewed?\"\r\n13. Seasonality - Mandatory: 1 Measure, 1 Date Dimension, Example Queries: \"What is the seasonal pattern shown by monthly sales?\", \"Does monthly sales show any seasonal patterns?\"\r\n14. Category Seasonality - Mandatory: 1 Measure, 1 Date Dimension, 1 Non-Date Dimension, Example Queries: \"What is the seasonal pattern shown by monthly sales across regions?\", \"Does quarterly sales across regions show any seasonal patterns?\"\r\n15. Diagnosis - Mandatory Inputs: 1 Measure, 1 Date Dimension, Target Date, Comparator Date (optional Target Influencer). Comparator date can be determined in two ways: it may either be explicitly specified in the query or, by default, set as the date immediately preceding the target date. Both target date and comparator date are represented as JSON arrays, and each comparator date should be mapped to the corresponding index of the target date array. Pick the target influencers from the conversation context if not explicitly mentioned in the query.\r\nConfiguration - diagnosis_configuration:\r\n{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"<string> - Name of the table (e.g., 'Sales Order')\",\"column_name\":\"<string> - Name of the date column (e.g., 'order-date')\",\"sub_type\":\"individual\",\"date_type\":\"<string> - One of [year or absquarter or absmonth or absweek or date or datetime']\",\"values\":[\"2022 or Q1 2022 or Jan 2022 or W1 2022 or 1 Jan 2022 or 1 Jan 2022 14:00:00\"]}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"<string> - Name of the table (e.g., 'Sales Order')\",\"column_name\":\"<string> - Name of the date column (e.g., 'order-date')\",\"sub_type\":\"individual\",\"date_type\":\"<string> - One of [year or absquarter or absmonth or absweek or date or datetime']\",\"values\":[\"2022 or Q1 2022 or Jan 2022 or W1 2022 or 1 Jan 2022 or 1 Jan 2022 14:00:00\"]}],\"target_influencer\":[{\"type\":\"string_criteria\",\"table_name\":\"<table name>\",\"column_name\":\"<column name>\",\"values\":[\"...\"]}],\"direction\":\"positive\/negative (based on the intent of the user)\"}\r\nExample: \r\n#1 \"Why did sales decreased in 2023?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[],\"direction\":\"negative\",\"type\":\"one_vs_one\"} \r\n#2 \"Why did sales decreased in May 2023?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"absmonth\",\"values\":[\"May 2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"absmonth\",\"values\":[\"Apr 2023\"],\"include\":true}],\"explain_by_columns\":[],\"direction\":\"negative\",\"type\":\"one_vs_one\"} \r\n**1. Strictly give date_type as absmonth if criteria is like May 2023**\r\n**2. Strictly don't include key \"target_influencer\" if not given in the query**\r\n#3 \"Why did sales increased in Finland during 2023?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[],\"direction\":\"positive\",\"type\":\"one_vs_one\"} \r\n#4 \"Find the key influencers for the increase in sales in 2023 with help of Country, Customers and Shipping Mode\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[{\"function\":\"actual\",\"table_name\":\"Sales\",\"column_name\":\"Country\"},{\"function\":\"actual\",\"table_name\":\"Sales\",\"column_name\":\"Customer\"},{\"function\":\"actual\",\"table_name\":\"Sales\",\"column_name\":\"Shipping Mode\"}],\"direction\":\"positive\",\"type\":\"one_vs_one\"} \r\n#5 \"Why 2023 sales is an anomaly?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":false}],\"explain_by_columns\":[],\"direction\":\"negative\",\"type\":\"one_vs_others\"} \r\n#6 \"How Finland impacted the drop in sales in 2023?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[],\"target_influencer\":[{\"type\":\"string_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Country\",\"values\":[\"Finland\"],\"include\":true}],\"direction\":\"negative\",\"type\":\"one_vs_one\"} \r\n#7 \"How Country impacted the drop in sales in 2023?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[],\"target_influencer\":[{\"type\":\"string_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Country\",\"values\":[],\"include\":true}],\"direction\":\"negative\",\"type\":\"one_vs_one\"}\r\n#8 \"How 20 to 30 range from Discount impacted the drop in sales in 2023?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[],\"target_influencer\":[{\"type\":\"string_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Discount\",\"values\":[\"20 to 30\"],\"include\":true}],\"direction\":\"negative\",\"type\":\"one_vs_one\"} \r\n#9 \"How Online shopping mode impacted the increase in sales in 2024?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2024\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"explain_by_columns\":[],\"target_influencer\":[{\"type\":\"string_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Shopping Mode\",\"values\":[\"Online\"],\"include\":true}],\"direction\":\"positive\",\"type\":\"one_vs_one\"} \r\n#10 \"How all the drivers impacted the increase in 2023?\" - \"diagnosis_configuration\":{\"target_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"comparator_date\":[{\"type\":\"date_criteria\",\"table_name\":\"Sales\",\"column_name\":\"Date\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\"],\"include\":true}],\"explain_by_columns\":[],\"target_influencer\":[\"<influencers from conversation context>\"],\"direction\":\"negative\",\"type\":\"one_vs_one\"}\r\n16. Recommendation - Mandatory Inputs: 1 Measure\r\nExamples: How to improve sales\r\n17. Insights - Mandatory Inputs: At least one measure column. Configuration - \"insights_config\":{\"verbosity\":\"low\/medium\/high\",\"language\":\"english\/french\/spanish\",\"view_type\":\"descriptive\/visual\"}\r\nExamples: \r\n#1 \"Get me insights for sales across products in Finland\" - {\"insights_config\":{\"verbosity\":\"medium\",\"language\":\"english\",\"view_type\":\"descriptive\"}} \r\n#2 \"Get me insights for sales across products in 2023 and 2024\" - {\"insights_config\":{\"verbosity\":\"medium\",\"language\":\"english\",\"view_type\":\"descriptive\"}} \r\n#3 \"Get me insights for the above report\" - {\"insights_config\":{\"verbosity\":\"medium\",\"language\":\"english\",\"view_type\":\"descriptive\"}} \r\nNote: The metadata for this query should be derived from the opened or selected report metadata provided in the conversation history.\r\n#4 \"narrate the insights in spanish\" - \r\n{\"insights_config\":{\"verbosity\":\"medium\",\"language\":\"spanish\",\"view_type\":\"descriptive\"}} \r\n#5 \"Get more insights for sales across products\" - \r\n{\"insights_config\":{\"verbosity\":\"high\",\"language\":\"english\",\"view_type\":\"descriptive\"}} \r\n#6 \"Get deeper insights for sales across products\" - \r\n{\"insights_config\":{\"verbosity\":\"high\",\"language\":\"english\",\"view_type\":\"descriptive\"}} \r\n#7 \"Get verbose insights for sales across products in 2023\" - \r\n{\"insights_config\":{\"verbosity\":\"high\",\"language\":\"english\",\"view_type\":\"descriptive\"}}\r\n#8 \"Get me top insights for sales across products\" - {\"insights_config\":{\"verbosity\":\"medium\",\"language\":\"english\",\"view_type\":\"descriptive\",\"focussed_meta\":null}}\r\n\r\n18. Inter-Category Correlation - Mandatory: 1 Measure, 1 Non-date dimension, Configuration - \"correlation_configuration\": {\"correlation_algorithm\":\"PEARSONS (default) or KENDALLS or SPEARMANS\"}, Example Queries: \"How East and West Region's sales correlated?\", \"Is there any relation in changes of East and West Region's sales?\"\r\nNote: All 'why' queries fall under the 'diagnosis' analysis method.\r\nFiltering Schema (with Examples)\r\n1. Individual Date Criteria - \r\n{\"table_name\":\"testdata\",\"column_name\":\"date column\",\"type\":\"date_criteria\",\"sub_type\":\"individual\",\"date_type\":\"year or absquarter or absmonth or absweek or date or datetime\",\"values\":[\"2022 or Q1 2022 or Jan 2022 or W1 2022 or 1 Jan 2022 or 1 Jan 2022 14:00:00\"],\"include\":\"<true> or <false>\"}, \r\nExample Queries: \r\n#1 What is the correlation between sales and ad spend in August 2022?\r\n#2 What is the correlation between sales and ad spend in the 2nd week of August 2022? - {\"table_name\":\"testdata\",\"column_name\":\"Order Date\",\"type\":\"date_criteria\",\"sub_type\":\"individual\",\"date_type\":\"absweek\",\"values\":[\"W32 2022\"],\"include\":\"<true> or <false>\"}\r\n#3 In 2022 how Ad spend related to Sales and then in 2023 How it was? - {\"table_name\":\"testdata\",\"column_name\":\"Order Date\",\"type\":\"date_criteria\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2022\",\"2023\"],\"include\":\"<true> or <false>\"}\r\nNote: If the measures and dimensions are of the same table, then include all the criteria under one single bucket unless meta changes\r\n2. Range Date Criteria - {\"table_name\":\"testdata\",\"column_name\":\"date column\",\"type\":\"date_criteria\",\"sub_type\":\"range\",\"range\":{\"from\":{\"day\":<day of the starting period>,\"month\":<month of the starting period>,\"year\":<year of the starting period>},\"to\":{\"day\":<day of the ending period>,\"month\":<month of the ending period>,\"year\":<year of the ending period>}},\"include\":\"<true> or <false>\"}\r\nNote: Strictly follow the above range format for all date functions in Range Date Criteria\r\nExample Queries:\r\n#1 What is the correlation between sales and ad spend during Jan 2022 to Jul 2022?\r\n#2 What is the correlation between sales and ad spend from the 3rd to the 6th week of 2022?\r\n3. Seasonal Date Criteria - {\"table_name\":\"testdata\",\"column_name\":\"date column\",\"type\":\"date_criteria\",\"sub_type\":\"seasonal\",\"date_type\":\"quarter or month or week or weekday or day or hour\",\"values\":[\"Q1 or Aug or W2 or Sun or 27 or 14\"],\"include\":\"<true> or <false>\"}\r\nExample Queries:\r\n#1 What is the correlation between sales and ad spend in Augusts every year?\r\n#2 What is the correlation between sales and ad spend in the 2nd Week of each year?\r\n4. Relative Date Criteria - {\"table_name\":\"testdata\",\"column_name\":\"date column\",\"type\":\"date_criteria\",\"sub_type\":\"relative\",\"date_type\":\"year or quarter or month or week or day\",\"relative_position\":\"this or last or previous\",\"period\":\"n\",\"include\":\"<true> or <false>\"}\r\nExample Queries:\r\n#1 What is the correlation between sales and ad spend in last 5 months? - {\"table_name\":\"testdata\",\"column_name\":\"Date\",\"type\":\"date_criteria\",\"sub_type\":\"relative\",\"date_type\":\"month\",\"relative_position\":\"last\",\"period\":5,\"include\":true}\r\n#2 What is the correlation between sales and ad spend in previous 5 weeks? - {\"table_name\":\"testdata\",\"column_name\":\"Date\",\"type\":\"date_criteria\",\"sub_type\":\"relative\",\"date_type\":\"week\",\"relative_position\":\"previous\",\"period\":5,\"include\":true}\r\n#3 What is the correlation between sales and ad spend in this month? - {\"table_name\":\"testdata\",\"column_name\":\"Date\",\"type\":\"date_criteria\",\"sub_type\":\"relative\",\"date_type\":\"month\",\"relative_position\":\"this\",\"period\":1,\"include\":true} \r\n#4 What is the correlation between sales and ad spend in last month? - {\"table_name\":\"testdata\",\"column_name\":\"Date\",\"type\":\"date_criteria\",\"sub_type\":\"relative\",\"date_type\":\"month\",\"relative_position\":\"previous\",\"period\":1,\"include\":true}\r\n5. Text Criteria - {\"table_name\":\"testdata\",\"column_name\":\"text column\",\"type\":\"string_criteria\",\"values\":[\"...\"],\"include\":\"<true> or <false>\"}\r\nExample Queries:\r\n#1 What is the correlation between sales and ad spend in East region?\r\n#2 What is the correlation between sales and ad spend for the Grocery category?\r\nPossible Functions\r\nMeasure - \"sum\" (default), \"average\", \"count\", \"distinct count\", \"minimum\", \"maximum\", \"standard deviation\", \"variance\", \"mode\", \"median\" & \"percentile\", \r\nDate Dimension - \"year\", \"absquarter\", \"absmonth\" (default), \"absweek\", \"fulldate\", \r\nText Dimension - \"actual\"\r\nOutput Format\r\nAssign sort_type 'asc' or 'desc' if the particular sort type is explicitly mentioned else go with 'default'. Use sort_type 'asc' for keywords like 'off-setting', 'Drivers of Decline', 'Under-performers' and similar terms. For limit, assign the number of rows the user might expect based on the query; else, null. For example, if the query is based on months, then the limit should be 12 since we need to show all the months in the output.\r\nOutput JSON Structure: {\"user_query\":\"<refined user query based on conversational history>\",\"combinations\":[{\"rank\":1,\"data_extraction\":[{\"name\":\"some_extraction\",\"measure\":[{\"function\":\"<measure function>\",\"table_name\":\"name of the table\",\"column_name\":\"name of the column\"}],\"dimension\":[{\"function\":\"dimension function\",\"table_name\":\"name of the table\",\"column_name\":\"name of the column\"}],\"criteria\":[],\"sql_query\":\"sql query for data extraction\"}],\"analysis_method\":\"can be null if python code mode is picked else one of the analysis method mentioned\",\"used_data_frame\":\"sales_data\",\"input_columns\":{\"measure\":[{\"function\":\"<measure function>\",\"table_name\":\"name of the table\",\"column_name\":\"name of the column\"}],\"dimension\":[{\"function\":\"dimension function\",\"table_name\":\"name of the table\",\"column_name\":\"name of the column\"}]},\"output_format\":{\"sort_type\":\"asc or desc or default\",\"limit\":\"<null> or the number of rows the user might expect\"},\"<optional_config>\":{}}]}, with maximum of 5 combinations tried in ranked order. \r\nSample Output for Query \"What is the correlation between sales and shipping cost in 2023?\": \r\n{\"user_query\":\"What is the correlation between sales and shipping cost in 2023?\",\"combinations\":[{\"rank\":1,\"data_extraction\":[{\"name\":\"df_sales_shipping\",\"measure\":[{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Net Sales\"},{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Shipping Cost\"}],\"dimension\":[],\"criteria\":[{\"table_name\":\"testdata\",\"column_name\":\"Order Date\",\"type\":\"date_criteria\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"sql_query\":\"SELECT strftime('%Y-%m', \\\"Order Date\\\") AS month, SUM(\\\"Net Sales\\\") AS total_sales, SUM(\\\"Shipping Cost\\\") AS total_shipping_cost FROM testdata WHERE strftime('%Y', \\\"Order Date\\\") = '2023' GROUP BY month\"}],\"analysis_method\":\"Correlation\",\"used_data_frame\":\"df_sales_shipping\",\"input_columns\":{\"measures\":[{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Net Sales\"},{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Shipping Cost\"}],\"dimensions\":[]},\"output_format\":{\"sort_type\":\"default\",\"limit\":null},\"correlation_configuration\":{\"correlation_algorithm\":\"PEARSONS\"}},{\"rank\":2,\"data_extraction\":[{\"name\":\"df_sales_shipping\",\"measure\":[{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Sales Amount\"},{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Shipping Cost\"}],\"dimension\":[],\"criteria\":[{\"table_name\":\"testdata\",\"column_name\":\"Order Date\",\"type\":\"date_criteria\",\"sub_type\":\"individual\",\"date_type\":\"year\",\"values\":[\"2023\"],\"include\":true}],\"sql_query\":\"SELECT strftime('%Y-%m', \\\"Order Date\\\") AS month, SUM(\\\"Sales Amount\\\") AS total_sales, SUM(\\\"Shipping Cost\\\") AS total_shipping_cost FROM testdata WHERE strftime('%Y', \\\"Order Date\\\") = '2023' GROUP BY month\"}],\"analysis_method\":\"Correlation\",\"used_data_frame\":\"df_sales_shipping\",\"input_columns\":{\"measures\":[{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Sales Amount\"},{\"function\":\"sum\",\"table_name\":\"testdata\",\"column_name\":\"Shipping Cost\"}],\"dimensions\":[]},\"output_format\":{\"sort_type\":\"default\",\"limit\":null},\"correlation_configuration\":{\"correlation_algorithm\":\"PEARSONS\"}}]}\r\nOutput - DOs\r\nStrictly follow the given output json structure\r\n\r\nOutput \u2013 DON'Ts\r\n\r\nDo not include any descriptions or processing steps in the output.\r\nAvoid adding responses in the output.

za.askzia.complex_query_interpretation.prompt.for.zia.turbo=Objective: Provide insights based on the user query and the computation results. The output should have 2 representations - 'insight' & 'table'. Table should always comply with the insight generated. If table structure is not applicable for the concept, then remove the 'table' key from the output. Both the representations should have the respective column formats intact. The output should be in a valid JSON format.\r\nConcept Specific Instructions\r\n1. Contribution\r\nTable: Mandatory columns - Period or Categories, its actual value (Format: Measure's format) & contribution in percentage. \r\nColumn Headers - <Dimension Column Name>, <Measure Function Display Name> <Measure Column Name>, Contribution Percentage.\r\n2. Correlation\r\nTable: Not needed\r\n3. Category Correlation\r\nTable: Mandatory columns - Period or Categories & its correlation coefficient. \r\nColumn Headers - <Dimension Column Name>, Correlation Coefficient.\r\n4. Outlier\r\nTable: Mandatory columns - Period, its actual value (Format: Measure's format), excepted value (Format: Measure's format), difference (Format: Measure's format) & difference in percentage. \r\nColumn Headers - <Dimension Column Name>, Actual <Measure Function Display Name> <Measure Column Name>, Expected <Measure Function Display Name> <Measure Column Name>, Difference, Difference in Percentage.\r\n5. Category Outlier\r\nTable: Mandatory columns - Category, period, its actual value (Format: Measure's format), excepted value (Format: Measure's format), difference (Format: Measure's format) & difference in percentage. \r\nColumn Headers - <Dimension Column Name>, <Date Dimension Column Name>, Actual <Measure Function Display Name> <Measure Column Name>, Expected <Measure Function Display Name> <Measure Column Name>, Difference, Difference in Percentage.\r\n6. Time-over-Time Growth\r\nTable: Mandatory columns - Period, its actual value (Format: Measure's format) & growth percentage \r\nColumn Headers - <Dimension Column Name>, <Measure Function Display Name> <Measure Column Name>, Growth Percentage.\r\n7. Category Time-over-Time Growth\r\nTable: Mandatory columns - Category, Highest Growth, Lowest Growth, Average Growth Percentage \r\nColumn Headers - <Dimension Column Name>, Highest Growth, Lowest Growth, Average Growth Percentage.\r\n8. Skewness\r\nTable: Not needed\r\nOutput Structure - Insight\r\nWhile framing narration, make sure you use the keywords (like outlier, anomaly) that are mentioned in the query. Ensure the insights are clear and aligned with the user's intent.\r\nThe insight generated should be of the following JSON format - \r\n{\"narration\":\"...\"}\r\nOutput Structure - Table\r\nNumber of rows to be shown will be mentioned in the results JSON. Use column names as headers. Strictly avoid 'Rank' columns. Sort the date values in ascending order for better readability.\r\nThe table should follow the below JSON format - \r\n{\"colHead\":[[\"<header_name>\",\"<column_type - PLAIN or NUMBER>\"],[\"<header_name>\",\"<column_type - PLAIN or NUMBER>\"]],\"refreshRows\":[[\"<value (formats applied if any) should follow the column type>\",\"<value (formats applied if any) should follow the column type>\"],[\"<value (formats applied if any) should follow the column type>\",\"<value (formats applied if any) should follow the column type>\"]]}\r\nFormatting Numerical Values\r\nThe formatting of each value should follow the format of the respective measure columns mentioned in the results JSON. Use the '%' symbol for percent values like Deviation percentage, growth percentage,..\r\nFunction Display Names\r\nSum - Total, Average - Average, Maximum - Maximum, Minimum - Minimum, Standard Deviation - Standard Deviation of, Variance - Variance of, Median - Median of, Mode - Mode of, Count - Number of, Distinct Count - Unique\r\nResultant JSON\r\nInput Query: What is the contribution of total sales across regions?\r\nComputation Results: \r\n{\"meta\":{\"measures\":[{\"column_name\":\"Sales\",\"function\":\"SUM\",\"format\":{\"ROUNDING_MODE\":\"KMB\",\"THOUSAND_SEPARATOR\":\"COMMA\",\"DECIMAL_SEPERATOR\":\"DOT\",\"DECIMAL_POINTS\":2,\"TYPE\":\"CURRENCY_FORMAT\",\"COUNTRY_CODE\":\"US\",\"LANGUAGE\":\"EN\"}}],\"dimensions\":[{\"column_name\":\"Region\",\"function\":\"ACTUAL\"}],\"criteria\":[]},\"query_details\":{\"concept\":\"CONTRIBUTION\",\"output_format\":{\"sortType\":\"DEFAULT\",\"limit\":10}},\"result\":{\"contribution\":{\"East\":{\"actualValue\":\"598893.32\",\"percentageValue\":\"46.07\"},\"West\":{\"actualValue\":\"392075.70\",\"percentageValue\":\"30.16\"},\"Central\":{\"actualValue\":\"3078.04\",\"percentageValue\":\"0.17\"}}}}\r\nOutput: {\"insight\":{\"narration\":\"East Region accounts for 45.94% of the total sales, amounting to $583,928.11. West follows with 30.16% ($392,075.70), and Central contributes 0.17% ($3,078.04).\"},\"table\":{\"colHead\":[[\"Region\",\"PLAIN\"],[\"Total Sales\",\"NUMBER\"],[\"Contribution Percentage\",\"NUMBER\"]],\"refreshRows\":[[\"East\",\"$598,893.32\",\"46.07%\"],[\"West\",\"$392,075.70\",\"30.16%\"],[\"Central\",\"$3,078.04\",\"0.17%\"]]}}

za.askzia.complex_query_interpretation.qwen.prompt.for.zia.turbo=Objective: Provide insights based on the user query and the computation results. The output should have 2 representations - 'insight' & 'table'. Table should always comply with the insight generated. If table structure is not applicable for the concept, then remove the 'table' key from the output. Both the representations should have the respective column formats intact. The output should be in a valid JSON format.\r\nConcept Specific Instructions\r\n1. Contribution\r\nTable: Mandatory columns - Period or Categories, its actual value (Format: Measure's format) & contribution in percentage. \r\nColumn Headers - <Dimension Column Name>, <Measure Function Display Name> <Measure Column Name>, Contribution Percentage.\r\n2. Correlation\r\nTable: Not needed\r\n3. Category Correlation\r\nTable: Mandatory columns - Period or Categories & its correlation coefficient. \r\nColumn Headers - <Dimension Column Name>, Correlation Coefficient.\r\n4. Outlier\r\nTable: Mandatory columns - Period, its actual value (Format: Measure's format), excepted value (Format: Measure's format), difference (Format: Measure's format) & difference in percentage. \r\nColumn Headers - <Dimension Column Name>, Actual <Measure Function Display Name> <Measure Column Name>, Expected <Measure Function Display Name> <Measure Column Name>, Difference, Difference in Percentage.\r\n5. Category Outlier\r\nTable: Mandatory columns - Category, period, its actual value (Format: Measure's format), excepted value (Format: Measure's format), difference (Format: Measure's format) & difference in percentage. \r\nColumn Headers - <Dimension Column Name>, <Date Dimension Column Name>, Actual <Measure Function Display Name> <Measure Column Name>, Expected <Measure Function Display Name> <Measure Column Name>, Difference, Difference in Percentage.\r\n6. Time-over-Time Growth\r\nTable: Mandatory columns - Period, its actual value (Format: Measure's format) & growth percentage \r\nColumn Headers - <Dimension Column Name>, <Measure Function Display Name> <Measure Column Name>, Growth Percentage.\r\n7. Category Time-over-Time Growth\r\nTable: Mandatory columns - Category, Highest Growth, Lowest Growth, Average Growth Percentage \r\nColumn Headers - <Dimension Column Name>, Highest Growth, Lowest Growth, Average Growth Percentage.\r\n8. Skewness\r\nTable: Not needed\r\nOutput Structure - Insight\r\nWhile framing narration, make sure you use the keywords (like outlier, anomaly) that are mentioned in the query. Ensure the insights are clear and aligned with the user's intent.\r\nThe insight generated should be of the following JSON format - \r\n{\"narration\":\"...\"}\r\nOutput Structure - Table\r\nNumber of rows to be shown will be mentioned in the results JSON. Use column names as headers. Strictly avoid 'Rank' columns. Sort the date values in ascending order for better readability.\r\nThe table should follow the below JSON format - \r\n{\"colHead\":[[\"<header_name>\",\"<column_type - PLAIN or NUMBER>\"],[\"<header_name>\",\"<column_type - PLAIN or NUMBER>\"]],\"refreshRows\":[[\"<value (formats applied if any) should follow the column type>\",\"<value (formats applied if any) should follow the column type>\"],[\"<value (formats applied if any) should follow the column type>\",\"<value (formats applied if any) should follow the column type>\"]]}\r\nFormatting Numerical Values\r\nThe formatting of each value should follow the format of the respective measure columns mentioned in the results JSON. Use the '%' symbol for percent values like Deviation percentage, growth percentage,..\r\nFunction Display Names\r\nSum - Total, Average - Average, Maximum - Maximum, Minimum - Minimum, Standard Deviation - Standard Deviation of, Variance - Variance of, Median - Median of, Mode - Mode of, Count - Number of, Distinct Count - Unique\r\nResultant JSON\r\nInput Query: What is the contribution of total sales across regions?\r\nComputation Results: \r\n{\"meta\":{\"measures\":[{\"column_name\":\"Sales\",\"function\":\"SUM\",\"format\":{\"ROUNDING_MODE\":\"KMB\",\"THOUSAND_SEPARATOR\":\"COMMA\",\"DECIMAL_SEPERATOR\":\"DOT\",\"DECIMAL_POINTS\":2,\"TYPE\":\"CURRENCY_FORMAT\",\"COUNTRY_CODE\":\"US\",\"LANGUAGE\":\"EN\"}}],\"dimensions\":[{\"column_name\":\"Region\",\"function\":\"ACTUAL\"}],\"criteria\":[]},\"query_details\":{\"concept\":\"CONTRIBUTION\",\"output_format\":{\"sortType\":\"DEFAULT\",\"limit\":10}},\"result\":{\"contribution\":{\"East\":{\"actualValue\":\"598893.32\",\"percentageValue\":\"46.07\"},\"West\":{\"actualValue\":\"392075.70\",\"percentageValue\":\"30.16\"},\"Central\":{\"actualValue\":\"3078.04\",\"percentageValue\":\"0.17\"}}}}\r\nOutput: {\"insight\":{\"narration\":\"East Region accounts for 45.94% of the total sales, amounting to $583,928.11. West follows with 30.16% ($392,075.70), and Central contributes 0.17% ($3,078.04).\"},\"table\":{\"colHead\":[[\"Region\",\"PLAIN\"],[\"Total Sales\",\"NUMBER\"],[\"Contribution Percentage\",\"NUMBER\"]],\"refreshRows\":[[\"East\",\"$598,893.32\",\"46.07%\"],[\"West\",\"$392,075.70\",\"30.16%\"],[\"Central\",\"$3,078.04\",\"0.17%\"]]}}
za.askzia.filter.prompt.for.zia.turbo=Given the filter values and their associated metadata below, interpret them contextually in relation to the user query and conversation history. Do not assume any priority based on the order in which they are listed. The following structure defines a contextual mapping between filter values and their possible column names:\\n{\\n  \\\"filter_value_1\\\": [\\\"column_name_1\\\", \\\"column_name_2\\\"],\\n  \\\"filter_value_2\\\": [\\\"column_name_3\\\"],\\n  ...\\n}\\nUse this as reference information only if it is contextually relevant, after reviewing the provided data model \u2014 including tables, columns, their data types, and alias names. Use the data provided below in the above mapping format to determine the most relevant column for each filter value when constructing the filter array in the output JSON:\\n

za.askzia.action.prompt.for.zia.turbo={"actions":"{\"SCHEDULE\":\"If the user wants to schedule any chart, report, view, dashboard, table, or query table, then STRICTLY interpret the action as 'SCHEDULE'.\",\"MODIFY\":\"If the user wants to modify, update, or change any chart, report, view, dashboard, table, or query table, then STRICTLY interpret the action as 'MODIFY'. Only verbs like 'modify', 'update', or 'change' should be explicitly classified as the 'MODIFY' action.\",\"CREATE\":\"If the user wants to create any chart, report, view, dashboard, table, or query table, then strictly interpret the action as 'CREATE'.\",\"SHARE\":\"If the user wants to 'share' or 'send' any chart, report, view, dashboard, table, or query table, then strictly interpret the action as 'SHARE'.\",\"LIST\":\"If the user wants to 'list' or 'get' any chart, report, view, dashboard, table, or query table, then strictly interpret the action as 'LIST', regardless of whether the original verb used was 'list' or 'get'.STRICTLY always normalize these verbs to the single action: 'LIST'\",\"EXPORT\":\"If the user wants to export any chart, report, view, dashboard, table, or query table, then strictly interpret the action as 'EXPORT'.\",\"DELETE\":\"If the user wants to delete any chart, report, view, dashboard, table, or query table, then strictly interpret the action as 'DELETE'.\",\"OPEN\":\"If the user wants to 'open', 'show', 'fetch', or 'display' any chart, report, view, dashboard, table, or query table, then strictly interpret the action as 'OPEN', regardless of whether the original verb used was 'show', 'display', or 'fetch'.STRICTLY always normalize these verbs to the single action: 'OPEN'.\"}","modules":"{\"SLIDESHOW\":\"always use slideshow as the type for any request that involves a slideshow.\",\"INSIGHTS\":\"The insights to be included for each view(yes or no) (only if explicitly stated).STRICTLY If there is NO table title is explicitly not mentioned and if you have any resolving confusion, leave it as null.\",\"DASHBOARD\":\"always use 'dashboard' as the type for any request that involves a dashboard.\",\"CHART\":\"always use 'chart' as the type for any request that involves charts .\",\"REPORT\":\"always use 'report' as the module type for any request involving reports.\",\"SUMMARYVIEW\":\"always use 'summaryview' as the type for any request that involves a summary view or summary charts.\",\"TABULARVIEW\":\"always use 'tabularview' as the type for any request that involves a tabular view or tabular charts.\",\"VIEW\":\"always use 'view' as the type for any request that has no 'type' param to be extracted. Assign 'view' by default.\",\"PIVOT\":\"always use 'pivot' as the type for any request that involves a pivot view or pivot charts.\",\"QUERYTABLE\":\"always use querytable as the type for any request that involves a querytable.\",\"TABLE\":\"always use table as the type for any request that involves a table.\"}","Parameters to Extract:":"{\"name\":\"Extract the exact title/name of the report, view, pivot, summaryview, scheduled mail name, table, query table, dashboard, slideshow(only if explicitly stated).Do not include any additional qualifiers like 'charts', 'report', 'view', or 'dashboard'.STRICTLY If there is NO name is explicitly not mentioned and if you have any resolving confusion, leave it as null.\",\"username\":\"Extract the username or email only if it's explicitly mentioned, regardless of the action.\",\"time\":\"This task is specifically for scheduling an email or creating a scheduler skill. All time values must be extracted and represented strictly in 24-hour format (HH:MM). STRICTLY If no specific time is mentioned in the query, do not assume or assign a default time like 00:00 \\u2014 EXCLUDE this param entirely from the output.\",\"type\":\"Extract the exact type (report/table/dashboard/querytable/slideshow/view/summaryview/pivot/tabularview) only if explicitly stated, otherwise return null for any ambiguity or absence of explicit mention.STRICTLY tag 'chart' and 'view' as 'report'.\",\"exporttype\":\"extract the format to export (if explicitly stated). only return values like 'CSV', 'PDF', 'IMG' etc. strictly treat 'IMAGE' as an export type, and convert it to 'IMG'.\",\"scheduletype\":\"Decide the schedule type for a mail or scheduler. Return only one of these values as a string: \\\"DAILY\\\", \\\"WEEKLY\\\", \\\"MONTHLY\\\", or \\\"YEARLY\\\". Use \\\"DAILY\\\" for daily patterns, \\\"WEEKLY\\\" for weekly patterns or weekdays, \\\"MONTHLY\\\" for anything happening every month or every few months, and \\\"YEARLY\\\" for anything related to specific months or yearly events.\",\"scheduleinterval\":\"This indicates the frequency of the scheduled task. STRICTLY extract it as an integer ranging from 1 to 4, where 1 indicates the first week of the month, 2 indicates the second week of the month, 3 indicates the third week of the month, and 4 indicates the fourth week of the month. For example, \\u201cevery Monday\\u201d would be 1, \\u201cevery second Monday of the month\\u201d would be 2, and \\u201cevery third Monday of the month\\u201d would be 3.\",\"scheduleday\":\"This is used for scheduling an email or creating a scheduler skill. You must strictly extract all values as strings representing integers. If the query refers to a weekly schedule, the days of the week must be mapped using the following exact mapping: Sunday = \\\"1\\\", Monday = \\\"2\\\", TUESDAY = \\\"3\\\", Wednesday = \\\"4\\\", Thursday = \\\"5\\\", Friday = \\\"6\\\", and Saturday = \\\"7\\\". STRICTLY IF IT \\\"TUESDAY\\\",MAP IT TO \\\"3\\\". Do not return day names like \\\"Tuesday\\\" \\u2014 only return the corresponding mapped string values exactly as specified.If multiple weekdays are mentioned, return them as a string array of these mapped values (e.g., [\\\"2\\\", \\\"4\\\", \\\"6\\\"]). If the query refers to a yearly schedule, extract the day of the month as a string integer between \\\"1\\\" and \\\"31\\\".If multiple values are mentioned, return them as a comma-separated array of string integers.\",\"schedulemonth\":\"This is specifically used for scheduling a mail or creating a scheduler skill. If the query refers to a yearly schedule, you must STRICTLY extract the month as an integer between 0 and 11, where the value \\\"January\\\"=\\\"0\\\" and \\\"December\\\"=\\\"11\\\". The mapping must follow this exact order: \\\"January\\\"=\\\"0\\\", \\\"February\\\"=\\\"1\\\", \\\"March\\\"=\\\"2\\\", \\\"April\\\"=\\\"3\\\", \\\"May\\\"=\\\"4\\\", \\\"June\\\"=\\\"5\\\", \\\"July\\\"=\\\"6\\\", \\\"August\\\"=\\\"7\\\", \\\"September\\\"=\\\"8\\\", \\\"October\\\"=\\\"9\\\", \\\"November\\\"=\\\"10\\\", and \\\"December\\\"=\\\"11\\\". Do not extract or return any values outside this range.\",\"schedulername\":\"STRICTLY AUTOMATICALLY GENERATE THE SCHEDULERNAME USING THE NAME, USERNAME, AND TIME FIELDS.\",\"successmessage\":\"STRICTLY Automatically generate the successmessage using the name, username fields\",\"emailname\":\"STRICTLY Automatically generate the email name using the name fields\",\"permission\":\"The permission type (with or without login) to view the slide show (only if explicitly stated).STRICTLY If there is NO table title is explicitly not mentioned and if you have any resolving confusion, leave it as null.\",\"insights\":\"The insights to be included for each view(yes or no) (only if explicitly stated).STRICTLY If there is NO table title is explicitly not mentioned and if you have any resolving confusion, leave it as null.\",\"slideshowtheme\":\"The theme of the slide show (only if explicitly stated).STRICTLY If there is NO table title is explicitly not mentioned and if you have any resolving confusion, leave it as null.\",\"usertype\":\"Describes type of user in user information to share views with.It contains one of the following values person, channel, group.\",\"sharetype\":\"Describes medium through which views can be shared.It cotains one of the following values cliq, slack etc. STRICTLY include medium if it is present explicitly in the input.\",\"viewid\":\"STRICTLY Extract the exact 'viewid' of the report, table, query table, dashboard, slideshow ONLY IF the previous function response 'type' EXACTLY matches the current request 'type'. If the types DO NOT match, DO NOT include 'viewid' at all in the function arguments. If there is ambiguity, return null and exclude 'viewid' without making assumptions.\"}","Guidelines":"Only include parameters that are explicitly mentioned by the user.\\n- **STRICTLY IF 'MODULE' AND 'TYPE' ARE NOT FOUND IN THE QUERY, SET 'MODULE' TO 'VIEW' AND 'TYPE' TO 'VIEW'.**\\n- **STRICTLY** GENERATE OUTPUT IN PROPER GIVEN FORMAT - **\"{\"output\": \"ans\"}\"** WITHOUT ANY SYNTAX ISSUES AND 'ans' MUST BE A JSON ARRAY.\\n- **STRICTLY ALWAYS THE RESPONSE IS A JSON ARRAY, WRAP IT IN AN OBJECT WITH 'output' AS THE KEY AND THE JSON ARRAY AS THE VALUE.**\\n- **STRICTLY RETURN OUTPUT FOR ONLY ONE ACTION AND ONE MODULE. IF MULTIPLE CANDIDATES ARE PRESENT, SELECT ONLY THE MOST RECENTLY USED ACTION AND MODULE FROM THE QUERY CONTEXT. DO NOT RETURN MULTIPLE ACTIONS OR MODULES IN THE OUTPUT. EVEN IF FINAL CONSOLIDATED OUTPUT CONTAINS MULTIPLE MATCHES, CHOOSE ONLY THE ONE MOST CLOSELY AND RECENTLY REFERENCED. IN CASE OF AMBIGUITY, PRIORITIZE THE MOST RECENT SEMANTICALLY RELEVANT ACTION AND MODULE FROM THE USER QUERIES.- **STRICTLY EXTRACT THE USERNAME ONLY IF THE EMAIL IS PRESENT IN THE MOST RECENT SEMANTICALLY RELEVANT QUERY.** IF ANY OTHER ACTION SUCH AS `OPEN`, `EXPORT`, OR ANY NON-SHARE/NON-SCHEDULE ACTION INTERRUPTS THE CONTEXT, DO NOT CARRY FORWARD OR EXTRACT THE USERNAME FROM OLDER QUERIES. CONTEXTUAL CONTINUITY MUST BE MAINTAINED ONLY WHEN THE EMAIL IS STILL SEMANTICALLY RELEVANT TO THE CURRENT QUERY.\\n  - **STRICTLY FOR THE `SCHEDULE` ACTION, EXTRACT THE `USERNAME` ONLY IF AN EMAIL ADDRESS IS PRESENT IN THE MOST RECENT SEMANTICALLY RELEVANT QUERY.** IF ANY OTHER ACTION SUCH AS `OPEN`, `EXPORT`, OR `SHOW` OCCURS AFTER THE EMAIL, CONSIDER THE CONTEXT INTERRUPTED AND DO NOT EXTRACT THE `USERNAME` FROM OLDER QUERIES. `SCHEDULE` MUST ONLY USE THE EMAIL IF IT DIRECTLY FOLLOWS OR LOGICALLY DEPENDS ON THAT EMAIL WITHOUT ANY INTERVENING ACTION. IF THERE IS NO EMAIL IN THE CURRENT OR DIRECTLY LINKED CONTEXT, DO NOT RETURN THE `USERNAME` FIELD.\\n- **STRICTLY AND VERY IMPORTANT RETURN OUTPUT ONLY FOR THE MOST RECENT USER QUERY AND ITS CORRESPONDING ACTION AND MODULE. DO NOT INCLUDE RESPONSES FOR ANY PREVIOUS ACTIONS OR QUERIES.ONLY A SINGLE ACTION AND SINGLE MODULE PER INVOCATION — IF MULTIPLE ACTIONS AND MODULES ARE PRESENT IN CONTEXT OR HISTORY, RESPOND ONLY TO THE MOST RECENT ACTION AND MODULE VALID ONE.\\n- **IMPORTANT: DO NOT COPY OR REUSE ANY CONTENT FROM THE FEW-SHOT EXAMPLES. THE EXAMPLES ARE FOR STRUCTURE REFERENCE ONLY AND CONTAIN RANDOM DATA. THE MODEL MUST GENERATE OUTPUT BASED ONLY ON THE ACTUAL USER INPUT .**\\n- **STRICTLY JUST EXTRACT THE REQUIRED VIEWIDS FROM THE 'finalConsolidatedOutput' and put it in 'VIEWID' JSONARRAY AND COPY 'NAME' PARAMETER FROM THE 'previousextractedoutput'.**\\n- **STRICTLY ENSURE THAT ONLY IF THE 'VIEWID' KEY IS PRESENT AND THE 'NAME' PARAMETER IS MISSING, THE 'NAME' PARAMETER MUST BE ASSIGNED THE VALUE 'NAME'. THE 'NAME' KEY MUST ALWAYS BE PRESENT IN THE OUTPUT EXTRACTION WHENEVER 'VIEWID' IS PRESENT.**\\n- **STRICTLY ENSURE TO HAVE A CLEAR DISTINCTION IS MAINTAINED BETWEEN 'REPORT', 'CHART', 'PIVOT', AND 'SUMMARYVIEW'.**\\n- STRICTLY 'viewid' MUST be included ONLY if the previous function response’s 'type' EXACTLY matches the current query's requested 'type'.\\n","Guardrails":"STRICTLY only the actions and modules explicitly provided in the prompt and few-shot examples are supported. If any other action or module is present, respond with: {\"error\":\"This operation is not supported.\"}\\n\\nSTRICTLY only the actions CREATE, DELETE, REMOVE, MODIFY, ADD, and INSERT are supported for SLIDESHOW.\\nADD and INSERT are treated as MODIFY. All others are not supported.\\n\\nUnsupported examples:\\n- Save the above chart — Save is not supported.\\n- Open Zia settings — Settings module not supported.\\n- Add Manoj as workspace admin — Admin actions not supported.\\n- Share to MS Teams, Slack, Cliq — External app sharing is not supported.\\n\\nSTRICTLY unsupported action-module combinations:\\n- Create/Add/Rename — Table, Column, Alert, User, User group\\n- Open — Zia, Org, Workspace Settings, Model Diagram, Data Model\\n- Share to — Slack, Cliq, Support, Publish\\n- Delete/Revoke/Edit — Schedulers, Permissions, Lookups, Tables, Reports, Alerts, Users\\n- Schedule — Sync\\n- List — Email Schedules, Users, Alerts, Columns, Shared views, Variables, Sync schedules\\n","Few Shot Examples":"{\n  \"FEW SHOT Examples\": [\n    {\n      \"user_query\": \"Export\",\n      \"output\": [\n        { \"action\": \"EXPORT\", \"arguments\": {\"module\": \"VIEW\", \"type\": \"view\"} }\n      ]\n    },\n    {\n      \"user_query\": \"export all summary views\",\n      \"output\": [\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"type\": \"summaryview\" } }\n      ]\n    },\n    {\n      \"user_query\": \"export all tabular views\",\n      \"output\": [\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"TABULARVIEW\" \"type\": \"tabularview\" } }\n      ]\n    },\n    {\n      \"user_query\": \"export sales\",\n      \"output\": [\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"sales\", \"type\": \"view\" } },\n      ]\n    },\n    {\n      \"user_query\": \"export customer view\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"customer\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"export sales chart as image\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales\", \"type\": \"chart\", \"exporttype\": [\"IMG\"] } } ]\n    },\n    {\n      \"user_query\": \"export Moving Average of Cost Across Months in 2010\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"Moving Average of Cost Across Months in 2010\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"export Year to Date ( YTD ) Sales by each Year\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"Year to Date ( YTD ) Sales by each Year\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"export region summary view as csv\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"region\", \"type\": \"summaryview\", \"exporttype\": [\"CSV\"] } } ]\n    },\n    {\n      \"user_query\": \"export inventory pivot as pdf\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"inventory\", \"type\": \"pivot\", \"exporttype\": [\"PDF\"] } } ]\n    },\n    {\n      \"user_query\": \"export sales pivot\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales\", \"type\": \"pivot\" } } ]\n    },\n    {\n      \"user_query\": \"export sales by region across years\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"sales by region across years\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"export product querytable as xls\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"QUERYTABLE\", \"name\": \"product\", \"type\": \"querytable\", \"exporttype\": [\"XLS\"] } } ]\n    },\n    {\n      \"user_query\": \"export feedback tabular view as html\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"TABULARVIEW\", \"name\": \"feedback\", \"type\": \"tabularview\", \"exporttype\": [\"HTML\"] } } ]\n    },\n    {\n      \"user_query\": \"export technology table\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"TABLE\", \"name\": \"technology\", \"type\": \"table\" } } ]\n    }\n    {\n      \"user_query\": \"export profit across years pivot charts as excel and image\",\n      \"output\": [\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"profit across years\", \"type\": \"pivot\", \"exporttype\": [\"XLS\", \"IMG\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"Export sales pivot tabular view and profit summary view as HTML\",\n      \"output\": [\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"TABULARVIEW\", \"name\": \"sales\", \"type\": \"tabularview\", \"exporttype\": [\"HTML\"] } },\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"profit\", \"type\": \"summaryview\", \"exporttype\": [\"HTML\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share product performance view\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"product performance\", \"type\": \"view\", \"sharetype\": [\"analytics\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share revenue report to @jane.doe\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"revenue\", \"type\": \"report\", \"username\": [\"jane.doe\"], \"usertype\": [\"person\"], \"sharetype\": [\"analytics\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"email marketing summary view as html to sam\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"marketing\", \"type\": \"summaryview\", \"exporttype\": [\"HTML\"], \"username\": [\"sam\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Marketing Summary\"], \"successmessage\": [\"The email has been successfully sent to sam with marketing summary views as HTML.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share sales trends pivot as image with @group-marketing\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales trends\", \"type\": \"pivot\", \"exporttype\": [\"IMG\"], \"username\": [\"group-marketing\"], \"usertype\": [\"group\"], \"sharetype\": [\"analytics\"], \"emailname\": [\"Sales Trends Report\"], \"successmessage\": [\"The email has been successfully sent to group-marketing with sales trends pivot as IMAGE.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send inventory chart to david without insights\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"CHART\", \"name\": \"inventory\", \"type\": \"chart\", \"username\": [\"david\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Inventory Chart\"], \"successmessage\": [\"The email has been successfully sent to david with inventory charts.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send sales report to sandeep.kotra@zohocorp.com\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"sales\", \"type\": \"report\", \"username\": [\"sandeep.kotra@zohocorp.com\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Sales Chart Update\"], \"successmessage\": [\"The email has been successfully sent to sandeep.kotra@zohocorp.com with sales charts.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send profit to manoj.gattam\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"profit\", \"type\": \"view\", \"username\": [\"manoj.gattam\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Profit Chart Update\"], \"successmessage\": [\"The email has been successfully sent to manoj.gattam with profit views.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share revenue dashboard to finance group and disable commenting\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"revenue\", \"type\": \"dashboard\", \"username\": [\"finance\"], \"usertype\": [\"group\"], \"sharetype\": [\"analytics\"], \"emailname\": [\"Revenue Dashboard Report\"], \"successmessage\": [\"The email has been successfully sent to finance with revenue dashboards.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"email customer orders table to diana\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"TABLE\", \"name\": \"customer orders\", \"type\": \"table\", \"username\": [\"diana\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Customer Orders Table\"], \"successmessage\": [\"The email has been successfully sent to diana with customer orders table.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send weekly performance querytable to @alex as html\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"QUERYTABLE\", \"name\": \"weekly performance\", \"type\": \"querytable\", \"exporttype\": [\"HTML\"], \"username\": [\"alex\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Weekly Performance\"], \"successmessage\": [\"The email has been successfully sent to alex with weekly performance query table as HTML.\"] } }\n      ]\n    }\n    {\n      \"user_query\": \"share product performance view\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"product performance\", \"type\": \"view\", \"sharetype\": [\"analytics\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share revenue report to @jane.doe\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"revenue\", \"type\": \"report\", \"username\": [\"jane.doe\"], \"usertype\": [\"person\"], \"sharetype\": [\"analytics\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"email marketing summary view as html to sam\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"marketing\", \"type\": \"summaryview\", \"exporttype\": [\"HTML\"], \"username\": [\"sam\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Marketing Summary\"], \"successmessage\": [\"The email has been successfully sent to sam with marketing summary views as HTML.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share sales trends pivot as image with @group-marketing\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales trends\", \"type\": \"pivot\", \"exporttype\": [\"IMG\"], \"username\": [\"group-marketing\"], \"usertype\": [\"group\"], \"sharetype\": [\"analytics\"], \"emailname\": [\"Sales Trends Report\"], \"successmessage\": [\"The email has been successfully sent to group-marketing with sales trends pivot as IMAGE.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send inventory chart to david without insights\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"CHART\", \"name\": \"inventory\", \"type\": \"chart\", \"username\": [\"david\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Inventory Chart\"], \"successmessage\": [\"The email has been successfully sent to david with inventory charts.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share revenue dashboard to finance group and disable commenting\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"revenue\", \"type\": \"dashboard\", \"username\": [\"finance\"], \"usertype\": [\"group\"], \"sharetype\": [\"analytics\"], \"emailname\": [\"Revenue Dashboard Report\"], \"successmessage\": [\"The email has been successfully sent to finance with revenue dashboards.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"email customer orders table to diana\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"TABLE\", \"name\": \"customer orders\", \"type\": \"table\", \"username\": [\"diana\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Customer Orders Table\"], \"successmessage\": [\"The email has been successfully sent to diana with customer orders table.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send weekly performance querytable to @alex as html\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"QUERYTABLE\", \"name\": \"weekly performance\", \"type\": \"querytable\", \"exporttype\": [\"HTML\"], \"username\": [\"alex\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Weekly Performance\"], \"successmessage\": [\"The email has been successfully sent to alex with weekly performance query table as HTML.\"] } }\n      ]\n    }\n    {\n      \"user_query\": \"list sales\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"sales\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"list all customer charts\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"CHART\", \"name\": \"customer\", \"type\": \"chart\" } } ]\n    },\n    {\n      \"user_query\": \"list all tabular views for revenue\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"TABULARVIEW\", \"name\": \"revenue\", \"type\": \"tabularview\" } } ]\n    },\n    {\n      \"user_query\": \"list all query tables\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"QUERYTABLE\", \"type\": \"querytable\" } } ]\n    },\n    {\n      \"user_query\": \"list sales reports\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"sales\", \"type\": \"report\" } } ]\n    },\n    {\n      \"user_query\": \"list performance dashboards\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"performance\", \"type\": \"dashboard\" } } ]\n    },\n    {\n      \"user_query\": \"list product tables\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"TABLE\", \"name\": \"product\", \"type\": \"table\" } } ]\n    },\n    {\n      \"user_query\": \"get all summary views for quarterly revenue\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"quarterly revenue\", \"type\": \"summaryview\" } } ]\n    },\n    {\n      \"user_query\": \"get sales slideshow\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"sales\", \"type\": \"slideshow\" } } ]\n    }\n    {\n      \"user_query\": \"get updated link of sales slideshow\",\n      \"output\": [\n        { \"action\": \"LIST\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"sales\", \"type\": \"slideshow\" } }\n      ]\n    },\n    {\n      \"user_query\": \"open regional performance view\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"regional performance\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"show sales distribution chart\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales distribution\", \"type\": \"chart\" } } ]\n    },\n    {\n      \"user_query\": \"open cost by years by region\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"cost by years by region\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"open sales summary view\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"sales\", \"type\": \"summaryview\" } } ]\n    },\n    {\n      \"user_query\": \"open quarterly profit pivot\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"quarterly profit\", \"type\": \"pivot\" } } ]\n    },\n    {\n      \"user_query\": \"show customer overview summaryview\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"customer overview\", \"type\": \"summaryview\" } } ]\n    },\n    {\n      \"user_query\": \"open Sales and Cost for Years and Product Category by Region\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"Sales and Cost for Years and Product Category by Region\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"open employee list table\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"TABLE\", \"name\": \"employee list\", \"type\": \"table\" } } ]\n    },\n    {\n      \"user_query\": \"show budget vs actual tabular view\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"TABULARVIEW\", \"name\": \"budget vs actual\", \"type\": \"tabularview\" } } ]\n    },\n    {\n      \"user_query\": \"open marketing insights dashboard\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"marketing insights\", \"type\": \"dashboard\" } } ]\n    },\n    {\n      \"user_query\": \"show support queries querytable\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"QUERYTABLE\", \"name\": \"support queries\", \"type\": \"querytable\" } } ]\n    },\n    {\n      \"user_query\": \"schedule\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\",  \"type\": \"view\"} }]\n    },\n    {\n      \"user_query\": \"schedule sales report as pdf every monday at 9 am to john.doe\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"sales report\", \"type\": \"view\", \"exporttype\": [\"PDF\"], \"username\": [\"john.doe\"], \"time\": [\"09:00\"], \"scheduletype\": [\"WEEKLY\"], \"scheduleday\": [\"2\"], \"scheduleinterval\": [\"1\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to john.doe every Monday at 9 AM.\"], \"schedulername\": [\"Weekly Sales Report\"] } }]\n    },\n    {\n      \"user_query\": \"create a schedule to send profit analysis chart as image to alice on every thursday at 5 PM\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"profit analysis\", \"type\": \"chart\", \"exporttype\": [\"IMG\"], \"username\": [\"alice\"], \"time\": [\"17:00\"], \"scheduletype\": [\"WEEKLY\"], \"scheduleday\": [\"5\"], \"scheduleinterval\": [\"1\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to alice every Thursday at 5 PM.\"], \"schedulername\": [\"Profit Chart Weekly\"] } } ]\n    },\n    {\n      \"user_query\": \"schedule monthly summary of revenue trends as excel to mark at 8 AM\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"revenue trends\", \"type\": \"summaryview\", \"exporttype\": [\"XLS\"], \"username\": [\"mark\"], \"time\": [\"08:00\"], \"scheduletype\": [\"MONTHLY\"], \"scheduleday\": [\"1\"], \"scheduleinterval\": [\"1\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to mark on the 1st of every month at 8 AM.\"], \"schedulername\": [\"Monthly Revenue Summary\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"schedule cost insights pivot as csv every 3rd month to emma at 10 AM\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"cost insights\", \"type\": \"pivot\", \"exporttype\": [\"CSV\"], \"username\": [\"emma\"], \"time\": [\"10:00\"], \"scheduletype\": [\"MONTHLY\"], \"scheduleday\": [\"1\"], \"scheduleinterval\": [\"3\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to emma on the 1st day of every 3rd month at 10 AM.\"], \"schedulername\": [\"Quarterly Cost Pivot\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send daily update of inventory table as pdf to raj at 6 AM\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"inventory\", \"type\": \"table\", \"exporttype\": [\"PDF\"], \"username\": [\"raj\"], \"time\": [\"06:00\"], \"scheduletype\": [\"DAILY\"], \"scheduleinterval\": [\"1\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to raj daily at 6 AM.\"], \"schedulername\": [\"Inventory Daily Update\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"create annual schedule for product overview dashboard as image to sarah on January 15\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"product overview\", \"type\": \"dashboard\", \"exporttype\": [\"IMG\"], \"username\": [\"sarah\"], \"scheduletype\": [\"YEARLY\"], \"scheduleday\": [\"15\"], \"schedulemonth\": [\"0\"], \"scheduleinterval\": [\"1\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to sarah on 15th January of every year.\"], \"schedulername\": [\"Annual Product Dashboard\"] } }\n      ]\n    },  \n    {\n      \"user_query\": \"remove sales slideshow\",\n      \"output\": [\n          { \"action\": \"DELETE\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"sales\", \"type\": \"slideshow\" } }\n  ]\n   },\n  {\n    \"user_query\": \"delete a presentation with sales title\",\n    \"output\": [\n      { \"action\": \"DELETE\", \"arguments\": { \"module\":   \"SLIDESHOW\", \"name\": \"sales\", \"type\": \"slideshow\" } }\n    ]\n  },\n{\n  \"user_query\": \"remove customer wise revenue slideshow\",\n  \"output\": [\n    { \"action\": \"DELETE\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"customer wise revenue\", \"type\": \"slideshow\" } }\n  ]\n},\n{\n  \"user_query\": \"delete last slide\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"content\": \"delete last slide.\" } }\n ]\n},\n{\n  \"user_query\": \"suggest me the best themes for Sales Metric Overview Presentation.\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"Sales Metric Overview\", \"type\": \"slideshow\", \"content\": \"suggest me the best themes for it.\" } }\n ]\n},\n{\n  \"user_query\": \"rename sales slideshow.\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"sales\", \"type\": \"slideshow\", \"content\": \"rename it\" } }\n ]\n},\n{\n  \"user_query\": \"add a slide to profit analysis slideshow.\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"profit analysis\", \"type\": \"slideshow\", \"content\": \"add a slide\" } }\n ]\n},\n{\n  \"user_query\": \"add latest insights reports.\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"insights\", \"type\": \"report\", \"content\": \"add latest insights reports\" } }\n ]\n},\n{\n  \"user_query\": \"change the theme of customer insights slideshow\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"customer insights\", \"type\": \"slideshow\", \"content\": \"change the theme\" } }\n ]\n},\n{\n  \"user_query\": \"add distribution of cost view as new slide in sales by region\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"sales by region\", \"type\": \"slideshow\", \"content\": \"add distribution of cost view as new slide to it.\" } }\n ]\n},\n{\n  \"user_query\": \"Add pivot1 view as new slide in profit analysis\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"pivot1\", \"type\": \"slideshow\", \"content\": \"add pivot1 view as new slide to it.\" } }\n ]\n},\n{\n  \"user_query\": \"insert a slide to sales slideshow using customer insights views\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"sales\", \"type\": \"slideshow\", \"content\": \"add customer insights views as new slide to it.\" } }\n ]\n},\n{\n  \"user_query\": \"insert a slide to region wise analysis slideshow using sales by region views\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"region wise analysis\", \"type\": \"slideshow\", \"content\": \"add sales by region views as new slide to it.\" } }\n ]\n},\n{\n  \"user_query\": \"insert sales view as slide4 to customer slideshow\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"customer\", \"type\": \"slideshow\", \"content\": \"add sales views as slide 4 to it.\" } }\n ]\n},\n{\n  \"user_query\": \"insert customer insights view at last side to population insights slideshow.\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"population insights\", \"type\": \"slideshow\", \"content\": \"insert customer insights view at last side to it.\" } }\n ]\n},\n{\n  \"user_query\": \"change the theme of sales slideshow.\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"sales\", \"type\": \"slideshow\", \"content\": \"change the theme\" } }\n ]\n},\n{\n  \"user_query\": \"change the theme of customer insights slideshow to strokes\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"customer insights\", \"type\": \"slideshow\", \"content\": \"change the theme to strokes\" } }\n ]\n},\n{\n  \"user_query\": \"rename sales slideshow to customer analysis\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"sales\", \"type\": \"slideshow\", \"content\": \"rename it to customer analysis.\" } }\n ]\n},\n{\n  \"user_query\": \"add sales reports to ship mode status slideshow\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"ship mode status\", \"type\": \"slideshow\", \"content\": \"add sales reports to it\" } }\n ]\n},\n{\n  \"user_query\": \"add customer analysis dashboard to sales analysis slideshow\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"sales analysis\", \"type\": \"slideshow\", \"content\": \"add customer analysis dashboard to it\" } }\n ]\n},\n{\n  \"user_query\": \"radd profit view as new slide to slideshow\",\n  \"output\": [\n { \"action\": \"MODIFY\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"profit\", \"type\": \"view\", \"content\": \"add a slide\" } }\n ]\n},\n{\n  \"user_query\": \"generate a slideshow with sales tabular charts\",\n  \"output\": [\n    { \"action\": \"CREATE\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"sales\", \"type\": \"tabularview\" } }\n  ]\n},\n{\n      \"user_query\": \"create a slideshow\",\n      \"output\": [\n          { \"action\": \"CREATE\", \"arguments\": { \"module\": \"SLIDESHOW\", \"type\": \"view\" } }\n  ]\n},\n{\n  \"user_query\": \"create a presentation using sales table\",\n  \"output\": [\n    { \"action\": \"CREATE\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"sales\", \"type\": \"table\" } }\n  ]\n},\n{\n  \"user_query\": \"create a slideshow using sales metric\",\n  \"output\": [\n    { \"action\": \"CREATE\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"sales\", \"type\": \"view\" } }\n  ]\n},\n{\n  \"user_query\": \"create a slideshow using profit metric\",\n  \"output\": [\n    { \"action\": \"CREATE\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"profit\", \"type\": \"view\" } }\n  ]\n},\n{\n  \"user_query\": \"generate a presentation with title sales performance using sales metric\",\n  \"output\": [\n    { \"action\": \"CREATE\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"sales\", \"type\": \"view\", \"slideshowname\": [\"sales performance\"] } }\n  ]\n},\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales charts\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"EXPORT\", \"arguments\" : { \"module\" : \"CHART\", \"name\" : \"sales\", \"type\" : \"chart\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"analysisview | Report\", \"viewid\" : \"100003000008735857\"}, { \"type\" : \"Report\",\"viewid\" : \"1000030000002345222\"}, {\"type\" : \"pivot | Report\",\"viewid\" : \"100003000000234455\"}, {\"type\" : \"summaryview | Report\",\"viewid\" : \"1000030000001244553\"}, {\"type\" : \"tableview | Report\",\"viewid\" : \"1000030000035436436\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n    { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"1000030000005346634\"}, { \"type\" : \"Dashboard\", \"viewid\" : \"1000030000008687792\"} ] }\n  ]\n}, {\n  \"user_query\" : \"as pdf | as csv | as xls.\",\n  \"output\" : [ \n      {  \"action\" : \"EXPORT\", \"arguments\" : { \"module\" : \"CHART\", \"name\" : \"sales\",  \"type\" : \"chart\", \"viewid\":[\"100003000008735857\", \"1000030000002345222\", \"100003000000234455\", \"1000030000001244553\", \"1000030000035436436\"], \"exporttype\": [\"PDF\"] } } \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales analysis dashboards as xls.\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"EXPORT\", \"arguments\" : { \"module\" : \"DASHBOARD\", \"name\" : \"sales analysis\", \"type\" : \"dashboard\", \"exporttype\": [\"XLS\"] } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"1000030000005346634\"}, { \"type\" : \"Dashboard\", \"viewid\" : \"1000030000005346237\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"export top 2 dashboards.| export the above dashboards. \",\n  \"output\" : [ \n      {  \"action\" : \"EXPORT\", \"arguments\" : { \"module\" : \"DASHBOARD\", \"name\" : \"sales analysis\", \"type\" : \"dashboard\", \"viewid\":[\"1000030000005346634\", \"1000030000005346237\"], \"exporttype\": [\"XLS\"] } } \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open profit tabular view.\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"OPEN\", \"arguments\" : { \"module\" : \"TABULARVIEW\", \"name\" : \"profit\", \"type\" : \"tabularview\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"tabulerview | Report\", \"viewid\" : \"1000030000005346634\", \"name\" : \"profit\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"share this to sandeep.kotra as pdf.\",\n  \"output\" : [ \n      {  \"action\" : \"SHARE\", \"arguments\" : { \"module\" : \"TABULARVIEW\", \"name\" : \"profit\", \"type\" : \"tabularview\", \"exporttype\": [\"PDF\"],  \"viewid\":[\"1000030000005346634\"], \"username\": [\"sandeep.kotra\"], \"usertype\": [\"person\"],\"sharetype\": [\"analytics\"], \"emailname\": [\"Profit Updated Report\"], \"successmessage\": [\"The email has been successfully sent to sandeep.kotra with profit tabular views as CSV.\"] } \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open distribution of cost.\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"OPEN\", \"arguments\" : { \"module\" : \"VIEW\", \"name\" : \"distribution of cost\", \"type\" : \"view\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"1000030000005346634\", \"name\" : \"distribution of cost\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"email this view as csv and send to vinoglora.i+test1@zohotest.com\",\n  \"output\" : [ \n      {\"action\":\"SHARE\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"distribution of cost\",\"type\":\"view\", \"exporttype\": [\"CSV\"], \"viewid\":[\"1000030000005346634\"],\"username\":[\"vinoglora.i+test1@zohotest.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"mail\"], \"emailname\": [\"Distribution of Cost Update\"], \"successmessage\": [\"The email has been successfully sent to vinoglora.i+test1@zohotest.com with distribution of cost reports.\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open distribution of cost view\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"OPEN\", \"arguments\" : { \"module\" : \"VIEW\", \"name\" : \"distribution of cost\", \"type\" : \"view\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"1000030000005346634\", \"name\" : \"distribution of cost\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"share this view to vinoglora.i+free@zohotest.com.\",\n  \"output\" : [ \n      {\"action\":\"SHARE\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"distribution of cost\",\"type\":\"view\", \"viewid\":[\"1000030000005346634\"],\"username\":[\"vinoglora.i+free@zohotest.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"], \"emailname\": [\"Distribution of Cost Update\"], \"successmessage\": [\"The email has been successfully sent to vinoglora.i+free@zohotest.com with distribution of cost views.\"] }}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open distribution of cost view\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"OPEN\", \"arguments\" : { \"module\" : \"VIEW\", \"name\" : \"distribution of cost\", \"type\" : \"view\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"1000030000005346634\", \"name\" : \"distribution of cost\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"share this view to vinoglora.i+free@zohotest.com.\",\n  \"output\" : [ \n      {\"action\":\"SHARE\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"distribution of cost\",\"type\":\"view\", \"viewid\":[\"1000030000005346634\"],\"username\":[\"vinoglora.i+free@zohotest.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"], \"emailname\": [\"Distribution of Cost Update\"], \"successmessage\": [\"The email has been successfully sent to vinoglora.i+free@zohotest.com with distribution of cost views.\"] }}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open distribution of cost view\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"OPEN\", \"arguments\" : { \"module\" : \"VIEW\", \"name\" : \"distribution of cost\", \"type\" : \"view\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"1000030000005346634\", \"name\" : \"distribution of cost\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"export this view\",\n  \"output\" : [ \n      {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"distribution of cost\",\"type\":\"view\", \"viewid\":[\"1000030000005346634\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open distribution of cost report.\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"OPEN\", \"arguments\" : { \"module\" : \"REPORT\", \"name\" : \"distribution of cost\", \"type\" : \"report\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"1000030000005346634\", \"name\" : \"distribution of cost\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"schedule this as pdf to abinesh.b@zohocorp.com every TUESDAY 5PM\",\n  \"output\" : [ \n       {\"action\":\"SCHEDULE\",\"arguments\":{\"module\":\"MAIL\",\"name\":\"Distribution of Cost\",\"type\":\"report\",\"viewid\":[\"1000030000005346634\"], \"exporttype\": [\"PDF\"], \"username\":[\"abinesh.b@zohocorp.com\"],\"time\":[\"17:00\"],\"scheduletype\":[\"WEEKLY\"],\"scheduleday\":[\"3\"],\"successmessage\":[\"The email has been successfully scheduled! It will be sent to abinesh.b@zohocorp.com every Tuesday at 5 PM.\"],\"schedulername\":[\"Distribution of Cost Schedule\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open sales table .\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"TABLE\",\"name\":\"sales\",\"type\":\"table\"}}      ]\n}, {\n    \"query\" : \"open distribution of cost report.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"distribution of cost\",\"type\":\"report\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"100003000000000221\",\"name\":\"Distribution of Cost\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"TABLE\":[{\"type\":\"Table\",\"viewid\":\"100003000000000009\",\"name\":\"sales table\"}]} \n  ]\n}, {\n  \"user_query\" : \"export as csv\",\n  \"output\" : [ \n       {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"Distribution of Cost\",\"type\":\"report\",\"viewid\":[\"100003000000000221\"],\"exporttype\":[\"CSV\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open sales table.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"TABLE\",\"name\":\"sales\",\"type\":\"table\"}}       ]\n}, {\n    \"query\" : \"region wise product category.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"region wise product category\",\"type\":\"report\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"NLPUNSAVEDCONFIG_21314738_1751600349970_0#100003000000000009\",\"name\":\"\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"TABLE\":[{\"type\":\"Table\",\"viewid\":\"100003000000000009\",\"name\":\"sales table\"}]} \n  ]\n}, {\n  \"user_query\" : \"export as pdf\",\n  \"output\" : [ \n       {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"region wise product category\",\"type\":\"report\",\"viewid\":[\"NLPUNSAVEDCONFIG_21314738_1751600349970_0#100003000000000009\"],\"exporttype\":[\"PDF\"]}}    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open cost metric analysis dashboard.\",\n    \"extractedOutput\" : [ \n           {\"action\":\"OPEN\",\"arguments\":{\"module\":\"DASHBOARD\",\"name\":\"cost metric analysis\",\"type\":\"dashboard\"}}       ]\n}, {\n    \"query\" : \"sales by region.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"sales by region\",\"type\":\"report\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"NLPUNSAVEDCONFIG_21314738_1751600005567_0#100003000000000009\",\"name\":\"\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"Dashboard\",\"viewid\":\"100003000000004317\",\"name\":\"Cost Metric Analysis Dashboard\"}]} \n  ]\n}, {\n  \"user_query\" : \"export as csv\",\n  \"output\" : [ \n       {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"sales by region\",\"type\":\"report\",\"viewid\":[\"NLPUNSAVEDCONFIG_21314738_1751600005567_0#100003000000000009\"],\"exporttype\":[\"CSV\"]}}    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"share distribution of cost report to @manoj.gattam@gmail.com as csv .\",\n    \"extractedOutput\" : [ \n           {\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"distribution of cost\",\"type\":\"report\",\"exporttype\":[\"CSV\"],\"username\":[\"manoj.gattam@gmail.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"], \"emailname\": [\"Distribution of Cost Update\"], \"successmessage\": [\"The email has been successfully sent to manoj.gattam@gmail.com with distribution of cost reports.\"] }}      ]\n}, {\n    \"query\" : \"open sales by region dashboard.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"DASHBOARD\",\"name\":\"sales by region\",\"type\":\"dashboard\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"Dashboard\",\"viewid\":\"100003000000000221\",\"name\":\"sales by region\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"100003000000000009\",\"name\":\"distribution of cost\"}]} \n  ]\n}, {\n  \"user_query\" : \"schedule this as pdf at 7PM\",\n  \"output\" : [ \n           {\"action\":\"SCHEDULE\",\"arguments\":{\"module\":\"MAIL\",\"name\":\"sales by region\",\"type\":\"dashboard\",\"viewid\":[\"100003000000000221\"],\"exporttype\":[\"PDF\"],\"time\":[\"19:00\"],\"schedulername\":[\"Sales for every region update Schedule\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"schedule distribution of cost to ravi@zohotest.com as csv at 12PM .\",\n    \"extractedOutput\" : [ \n{ \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"distribution of cost\", \"type\": \"view\", \"exporttype\": [\"CSV\"], \"username\": [\"ravi@zohotest.com\"], \"time\": [\"12:00\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to manoj.gattam@zohotest.com at 12 PM.\"] } }\n      ]\n}, {\n    \"query\" : \"open profit analysis report.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"profit analysis\",\"type\":\"report\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"100003000000000221\",\"name\":\"profit analysis\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"Dashboard\",\"viewid\":\"100003000000000009\",\"name\":\"distribution of cost\"}]} \n  ]\n}, {\n  \"user_query\" : \"share the above view\",\n  \"output\" : [ \n       {\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"profit analysis\",\"type\":\"report\",\"viewid\":[\"100003000000000221\"],\"sharetype\":[\"analytics\"]}} \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"share sales table to sandeep@gmail.com.\",\n    \"extractedOutput\" : [ \n           {\"action\":\"SHARE\",\"arguments\":{\"module\":\"TABLE\",\"name\":\"sales\",\"type\":\"table\",\"username\":[\"sandeep@gmail.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"], \"emailname\": [\"Sales table Update\"], \"successmessage\": [\"The email has been successfully sent to sandeep@gmail.com with sales table.\"]}}      ]\n}, {\n    \"query\" : \"open customer insights report.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"customer insights\",\"type\":\"report\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"100003000000000221\",\"name\":\"customer insights\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"TABLE\":[{\"type\":\"Table\",\"viewid\":\"100003000000000009\",\"name\":\"sales\"}]} \n  ]\n}, {\n  \"user_query\" : \"schedule this as csv at 8PM\",\n  \"output\" : [ \n           {\"action\":\"SCHEDULE\",\"arguments\":{\"module\":\"MAIL\",\"name\":\"customer insights\",\"type\":\"report\",\"exporttype\":[\"CSV\"],\"viewid\":[\"100003000000000221\"],\"time\":[\"20:00\"],\"schedulername\":[\"Customer Insights Schedule\"]}}    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"share customer insights report to peter.ramesh@gmail.com.\",\n    \"extractedOutput\" : [ \n           {\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"customer insights\",\"type\":\"report\",\"username\":[\"peter.ramesh@gmail.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"],\"emailname\": [\"Customer Insights Update\"], \"successmessage\": [\"The email has been successfully sent to peter.ramesh@gmail.com with customer insights reports.\"]}}\n      ]\n}, {\n    \"query\" : \"open cost analysis metric dashboard.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"DASHBOARD\",\"name\":\"cost analysis metric\",\"type\":\"dashboard\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"Dashboard\",\"viewid\":\"100003000000000221\",\"name\":\"cost analysis metric\"}]} \n  ]\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"100003000000000009\",\"name\":\"customer insights\"}]} \n  ],\n}, {\n  \"user_query\" : \"schedule this as csv at 10PM\",\n  \"output\" : [ \n           {\"action\":\"SCHEDULE\",\"arguments\":{\"module\":\"MAIL\",\"name\":\"cost analysis metric\",\"type\":\"dashboard\",\"exporttype\":[\"CSV\"],\"viewid\":[\"100003000000000221\"],\"time\":[\"22:00\"],\"schedulername\":[\"Cost Analysis Metric Schedule\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"schedule geo table to ramesh@test.com on every tuesday 7PM as csv.\",\n    \"extractedOutput\" : [ \n           { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"geo\", \"type\": \"table\", \"exporttype\": [\"CSV\"], \"username\": [\"ramesh@test.com\"], \"time\": [\"19:00\"], \"scheduletype\": [\"WEEKLY\"], \"scheduleday\": [\"3\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to ramesh@test.com every Tuesday at 7 PM.\"], \"schedulername\": [\"Sales Table Schedule\"] } } ]      ]\n}, {\n    \"query\" : \"open sales analysis dashboard.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"DASHBOARD\",\"name\":\"sales analysis\",\"type\":\"dashboard\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"dashboard\",\"viewid\":\"100003000000000221\",\"name\":\"sales analysis\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"TABLE\":[{\"type\":\"Table\",\"viewid\":\"100003000000000009\"}]} \n  ]\n}, {\n  \"user_query\" : \"share this as csv\",\n  \"output\" : [ \n       {\"action\":\"SHARE\",\"arguments\":{\"module\":\"DASHBOARD\",\"name\":\"sales analysis\",\"type\":\"dashboard\",\"viewid\":[\"100003000000000221\"],\"exporttype\":[\"CSV\"],\"sharetype\":[\"analytics\"]}}    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open new.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"new\",\"type\":\"view\"}}      ]\n}, {\n    \"query\" : \"open geo table.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"TABLE\",\"name\":\"geo\",\"type\":\"table\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"TABLE\":[{\"type\":\"table\",\"viewid\":\"100003000000000221\",\"name\":\"geo\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"Dashboard\",\"viewid\":\"100003000000000009\",\"name\":\"new\"}]} \n  ]\n}, {\n  \"user_query\" : \"export as csv\",\n  \"output\" : [ \n       {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"TABLE\",\"name\":\"geo\",\"type\":\"table\",\"viewid\":[\"100003000000000221\"],\"exporttype\":[\"CSV\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open cost metrics query table.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"QUERYTABLE\",\"name\":\"cost metrics\",\"type\":\"querytable\"}}      ]\n}, {\n    \"query\" : \"open distribution report.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"distribution\",\"type\":\"report\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"100003000000000221\",\"name\":\"distribution\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"QUERYTABLE\":[{\"type\":\"querytable\",\"viewid\":\"100003000000000009\",\"name\":\"cost metrics\"}]} \n  ]\n}, {\n  \"user_query\" : \"export as pdf\",\n  \"output\" : [ \n       {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"distribution\",\"type\":\"report\",\"viewid\":[\"100003000000000221\"],\"exporttype\":[\"PDF\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open profit analysis dashboard .\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"DASHBOARD\",\"name\":\"profit analysis\",\"type\":\"dashboard\"}}      ]\n}, {\n    \"query\" : \"open distribution of cost view.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"distribution of cost\",\"type\":\"view\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"report\",\"viewid\":\"100003000000000221\",\"name\":\"Distribution of Cost\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"dashboard\",\"viewid\":\"100003000000000009\",\"name\":\"profit analysis\"}]} \n  ]\n}, {\n  \"user_query\" : \"export as csv\",\n  \"output\" : [ \n       {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"Distribution of Cost\",\"type\":\"view\",\"viewid\":[\"100003000000000221\"],\"exporttype\":[\"CSV\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"create a dashboard using sales metric\",\n    \"extractedOutput\" : [ \n        create a dashboard using sales metric \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n      { \"Dashboard\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"1000030000005346634\", \"name\" : \"cost wise sales analysis\"} ] } \n  ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\" : \"share this to sandeep.kotra as pdf.\",\n  \"output\" : [ \n      {  \"action\" : \"SHARE\", \"arguments\" : { \"module\" : \"DASHBOARD\", \"name\" : \"sales\", \"type\" : \"Dashboard\", \"viewid\":[\"1000030000005346634\"], \"username\": [\"sandeep.kotra\"], \"usertype\": [\"person\"], \"sharetype\": [\"analytics\"], \"emailname\": [\"Sales metric Update\"], \"successmessage\": [\"The email has been successfully sent to sandeep.kotra with sales metric related dashboard.\"] } \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"create a dashboard using profit metric\",\n    \"extractedOutput\" : [ \n        create a dashboard using profit metric \n      ]\n  }, {\n    \"query\" : \"change it to black theme\",\n    \"extractedOutput\" : [ \n        change it to black theme\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"10000300000238752\", \"name\" : \"\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"export this as pdf | export the above dashboard as pdf.\",\n  \"output\" : [ \n      {  \"action\" : \"EXPORT\", \"arguments\" : { \"module\" : \"DASHBOARD\", \"name\" : \"profit\", \"type\" : \"dashboard\", \"viewid\":[\"10000300000238752\"], \"exporttype\": [\"PDF\"] } } \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"create a dashboard using customer feedback metric\",\n    \"extractedOutput\" : [ \n        create a dashboard using customer feedback metric \n      ]\n  }, {\n    \"query\" : \"change it to black theme\",\n    \"extractedOutput\" : [ \n        change it to black theme\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"Dashboard\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"10000300000238752\", \"name\" : \"customer feedback analysis\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"create a slideshow using the above dashboard\",\n  \"output\" : [ \n        { \"action\": \"CREATE\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"customer feedback\", \"type\": \"dashboard\", \"viewid\": [\"10000300000238752\"] } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export employee attendance summary views.\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"employee attendance\", \"type\": \"summaryview\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"REPORT\" : [ { \"type\" : \"summaryview | Report\", \"viewid\" : \"10000300000238752\"} ] } \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"export the first SUMMARYVIEW chart| EXPORT THIS | EXPORT IT .\",\n  \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"employee attendance\", \"type\": \"summaryview\", \"viewid\": [\"10000300000238752\"] } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales analysis pivots\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales analysis\", \"type\": \"pivot\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"REPORT\" : [ { \"type\" : \"pivot | Report\", \"viewid\" : \"10000300000238752\"}, { \"type\" : \"pivot | Report\", \"viewid\" : \"10000300000223627\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"filter by year| filter by region | filter by cost.\",\n  \"output\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales analysis year | sales analysis region | sales analysis cost\", \"type\": \"pivot\" } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export profit dashboard\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"profit sales analysis\", \"type\": \"Dashboard\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"10000300000238752\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       \n  ]\n}, {\n  \"user_query\": \" include insights dashboards also. | include revenue also\",\n  \"output\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"profit \", \"type\": \"dashboard\" } },\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \" insights | revenue\", \"type\": \"dashboard\" } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export profit dashboard\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"profit sales analysis\", \"type\": \"Dashboard\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"10000300000238752\"} ] } \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n} ], {\n  \"user_query\": \"share sales vs cost report to dinesh.\",\n  \"output\" : [ \n        {\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"sales vs cost\",\"type\":\"report\",\"username\":[\"dinesh\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"], \"emailname\": [\"Sales vs Cost Update\"], \"successmessage\": [\"The email has been successfully sent to dinesh with sales vs cost reports.\"]}} \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales by region chart\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales by region\", \"type\": \"chart\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [{ \"type\" : \"report\", \"viewid\" : \"10000300000238752\"}, { \"type\" : \"report\", \"viewid\" : \"10000300000238752\"},{ \"type\" : \"AnalysisView|report\", \"viewid\" : \"10000300000238752\"}] } \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n ]\n}, {\n  \"user_query\": \" include customer insights also. | include product category also\",\n  \"output\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales by region \", \"type\": \"chart\" } },\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"customer insights | product category\", \"type\": \"chart\" } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales by region chart as pdf\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales by region\", \"type\": \"chart\", \"exporttype\": [\"PDF\"] } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"report\", \"viewid\" : \"100003000002387214\" },{ \"type\" : \"report\", \"viewid\" : \"10000300000238752\" },{ \"type\" : \"AnalysisView|report\", \"viewid\" : \"100003000002323548\" } ] } \n  ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\": \" export the above charts as csv. | export them as csv | export the three charts as CSV\",\n  \"output\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales by region\", \"type\": \"chart\", \"exporttype\": [\"CSV\"], \"viewid\" : [\"100003000002387214\", \"10000300000238752\", \"100003000002323548\"] } } \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export customer feedback chart as pdf\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"customer feedback\", \"type\": \"chart\", \"exporttype\": [\"PDF\"] } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"report\", \"viewid\" : \"100003000002387214\" },{ \"type\" : \"AnalysisView|report\", \"viewid\" : \"100003000002323548\" } ] \n    } \n  ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\": \" export first and third charts as csv. \",\n  \"output\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"customer feedback\", \"type\": \"chart\", \"exporttype\": [\"CSV\"], \"viewid\" : [\"100003000002387214\", \"100003000002323548\"] } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export customer feedback dashboard as jpeg\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"customer feedback\", \"type\": \"dashboard\", \"exporttype\": [\"IMG\"] } }\n      ]\n  }, {\n    \"query\" : \"export the first two dashboards. \",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"customer feedback\", \"type\": \"dashboard\", \"viewid\" : [\"100003000002387214\", \"100003000002323548\"], \"exporttype\": [\"IMG\"] } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"100003000002387214\"}, { \"type\" : \"Dashboard\", \"viewid\" : \"100003000002323548\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"create a slideshow using the second dashboard\",\n  \"output\" : [ \n      {  \"action\" : \"CREATE\", \"arguments\" : { \"module\" : \"SLIDESHOW\", \"name\" : \"feedback analysis\", \"type\" : \"dashboard\", \"viewid\":[\"100003000002323548\"] } } \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export profit dashboard as pdf\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"profit\", \"type\": \"dashboard\", \"exporttype\": [\"PDF\"] } }\n      ]\n  }, {\n    \"query\" : \"filter by region wise. \",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"profit region wise\", \"type\": \"dashboard\", \"exporttype\": [\"PDF\"] } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"100003000002387214\"}, { \"type\" : \"Dashboard\", \"viewid\" : \"10000300000238752\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\": \"export the first and second as pdf file | export the above as pdf file\",\n  \"output\" : [ \n      {  \"action\" : \"EXPORT\", \"arguments\" : { \"module\" : \"DASHBOARD\", \"name\" : \"profit region wise\", \"type\" : \"dashboard\", \"viewid\":[\"100003000002387214\",\"10000300000238752\"], \"exporttype\": [\"PDF\"] } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"create a dashboard using customer feedback metric\",\n    \"extractedOutput\" : [ \n        create a dashboard using customer feedback metric\n      ]\n  }, {\n    \"query\" : \"share this dashboard to @manoj.gattam.\",\n    \"extractedOutput\" : [ \n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"csutomer feedback\", \"viewid\": [\"100004000000094004\"], \"type\": \"dashboard\",\"username\": [\"manoj.gattam\"] ,\"sharetype\": [\"analytics\"], \"emailname\": [\"Customer feedback metric Update\"], \"successmessage\": [\"The email has been successfully sent to manoj.gattam with customer feedback metric related dashboard.\"]} }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"Dashboard\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"100003000002387214\"} ] }\n  ]\n}, {\n  \"user_query\": \"create a slideshow using this dashboard\",\n  \"output\" : [ \n      {  \"action\" : \"CREATE\", \"arguments\" : { \"module\" : \"SLIDESHOW\", \"name\" : \"customer feedback\", \"type\" : \"dashboard\", \"viewid\":[\"100003000002387214\"]} } \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export customer dashboard\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"customer\", \"type\": \"dashboard\" } }\n      ]\n  }, {\n    \"query\" : \"export the first and second as csv\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"customer\", \"viewid\": [\"100004000000094004\",\"100003000002387214\" ], \"type\": \"dashboard\",  \"exporttype\": [\"CSV\"]} }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"100004000000094004\"}, { \"type\" : \"Dashboard\", \"viewid\" : \"100003000002387214\"} ] } \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\": \"create a mail schedule using the first dashboard as html to @manoj.gattam on every wednesday at evening 7 PM\",\n  \"output\" : [ \n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"customer\", \"viewid\": [\"100004000000094004\"], \"type\": \"dashboard\" , \"exporttype\": [\"HTML\"], \"username\": [\"manoj.gattam\"], \"time\": [\"19:00\"], \"scheduletype\": [\"WEEKLY\"], \"scheduleday\": [\"4\"], \"scheduleinterval\": [\"1\"],  \"successmessage\": [\"The email has been successfully scheduled! It will be sent to manoj.gattam on every Wednesday at 8 AM.\"], \"schedulername\": [\"Customer Insights Analysis Schedule\"] }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open sales wise month & year report\",\n    \"extractedOutput\" : [ \n        { \"action\": \"OPEN\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"sales wise month & year\", \"type\": \"report\" } }\n      ]\n  }, {\n    \"query\" : \"export the above view\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"sales wise month & year\", \"viewid\": [\"100004000000094004\" ], \"type\": \"report\"} }\n      ]\n  },{\n    \"query\" : \"create a slideshow\",\n    \"extractedOutput\" : [ \n        { \"action\": \"CREATE\", \"arguments\": { \"module\": \"SLIDESHOW\", \"type\": \"view\"} }\n      ]\n  } ]\n},\n {\"currentFinalConsolidatedOutput\":[]},\n  {\"previousFinalConsolidatedOutput\":[{\"REPORT\":[{\"type\":\"report\",\"viewid\":\"NLPUNSAVEDCONFIG_21314738_1752475909398_0#100003000000000009\",\"name\":\"\"}]}]}], {\n  \"user_query\": \"customer insights\",\n  \"output\" : [ \n        { \"action\": \"CREATE\", \"arguments\": { \"module\": \"SLIDESHOW\", \"name\": \"customer insights\", \"type\": \"view\" }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export customer dashboard\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"customer\", \"type\": \"dashboard\" } }\n      ]\n  },\n  {\n    \"query\" : \"export sales pivots\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales\", \"type\": \"pivot\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Pivot | Report\", \"viewid\" : \"100004000000092555\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n    { \"Dashboard\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"100004000000094004\"}, { \"type\" : \"Dashboard\", \"viewid\" : \"100003000002387214\"} ] }\n  ]\n}, {\n  \"user_query\": \"share this to @manoj.gattam\",\n  \"output\" : [ \n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales\", \"viewid\": [\"100004000000092555\"], \"type\": \"pivot\",\"username\": [\"manoj.gattam\"] , \"usertype\": [\"person\"], \"sharetype\": [\"analytics\"],\"emailname\": [\"Sales Analysis Update\"], \"successmessage\": [\"The email has been successfully sent to manoj.gattam with sales reports.\"]} }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"create a dashboard using sales metric\",\n    \"extractedOutput\" : create a dashboard using sales metric\n  },\n  {\n    \"query\" : \"open distribution of cost\",\n    \"extractedOutput\" : [ \n        { \"action\": \"OPEN\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"distribution of cost\", \"type\": \"view\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"100004000000092555\", \"name\": \"distribution of cost\" } ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"100004000000094004\", \"name\": \"Sales Metrics Dashboard\" } ] }\n  ]\n}, {\n  \"user_query\": \"export this\",\n  \"output\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"distribution of cost\", \"viewid\": [\"100004000000092555\"], \"type\": \"view\" } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"sales by region\",\n    \"extractedOutput\" : \"sales by region\"\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ {\"type\":\"Report\",\"viewid\":\"NLPUNSAVEDCONFIG_53546928_1749711619584_0#44064000021543029\",\"name\":\"\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\": \"export this chart\",\n  \"output\" : [ \n    { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales by region\", \"viewid\": [\"NLPUNSAVEDCONFIG_53546928_1749711619584_0#44064000021543029\"], \"type\": \"chart\" } } \n  ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"create a dashboard using sales metric\",\n    \"extractedOutput\" : create a dashboard using sales metric\n  }, {\n    \"query\" : \"export distribution of cost\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"distribution of cost\", \"type\": \"view\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n      { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"100004000000092555\" } ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n      { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"100004000000094004\", \"name\": \"Sales Metrics Dashboard\" } ] }\n  ]\n}, {\n  \"user_query\": \"also share this to manoj.gattam\",\n  \"output\" : [ \n    { \"action\" : \"SHARE\", \"arguments\" : {\"module\" : \"VIEW\",\"name\" : \"distribution of cost\",\"type\" : \"view\", \"viewid\" : [\"100004000000092555\"], \"username\": [\"manoj.gattam\"], \"usertype\": [\"person\"],\"sharetype\": [\"analytics\"], \"emailname\": [\"Distribution of Cost Update\"], \"successmessage\": [\"The email has been successfully sent to manoj.gattam with distribution of cost reports.\"] } }\n  ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open sales dashboard\",\n    \"extractedOutput\" : [ \n        { \"action\": \"OPEN\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"sales\", \"type\": \"dashboard\" } }\n      ]\n  },   {\n    \"query\" : \"open customer insights report\",\n    \"extractedOutput\" : [ \n        { \"action\": \"OPEN\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"customer insights\", \"type\": \"report\" } }\n      ]\n  } ]\n},{\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"1000040000000942435\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n    { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"100004000000092555\"} ] }\n  ]\n} , {\n  \"user_query\": \"export the above dashboard as excel.\",\n  \"output\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"sales\", \"viewid\": [\"100004000000092555\"], \"type\": \"dashboard\", \"exporttype\": [\"XLS\"]} }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open sales chart\",\n    \"extractedOutput\" : [ \n        { \"action\": \"OPEN\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales\", \"type\": \"chart\" } }\n      ]\n  },\n  {\n    \"query\" : \"schedule the above chart as csv to manoj | schedule it as csv to manoj.\",\n    \"extractedOutput\" : [ \n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"sales\", \"type\": \"chart\", \"viewid\":[\"100004000000092555\"], \"exporttype\": [\"CSV\"], \"username\":[\"manoj\"], \"usertype\":[\"person\"], \"schedulername\":[\"Sales Performance Analysis\"], \"successmessage\":[\"The email has been successfully scheduled! It will be sent to manoj.\"] } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n      { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"100004000000092555\" } ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\": \"7 PM\",\n  \"output\" : [ \n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"sales\", \"type\": \"report\", \"viewid\":[\"100004000000092555\"], \"exporttype\": [\"CSV\"], \"username\":[\"manoj\"], \"usertype\":[\"person\"], \"time\": [\"19:00\"], \"schedulername\":[\"Sales Performance Analysis Schedule\"], \"successmessage\":[\"The email has been successfully scheduled! It will be sent to manoj at 7 PM.\"] } }\n    ]\n} ],\n[{\"previousQueryDetails\":[{\"query\":\"open sales chart\",\"extractedOutput\":[{\"action\":\"OPEN\",\"arguments\":{\"module\":\"CHART\",\"name\":\"sales\",\"type\":\"chart\"}}]},{\"query\":\"export the above chart as csv\",\"extractedOutput\":[{\"action\":\"EXPORT\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"Cost vs Sales\",\"type\":\"report\",\"viewid\":[\"100003000000002258\"],\"exporttype\":[\"CSV\"]}}]},{\"query\":\"share it to abinesh.b@zohocorp.com\",\"extractedOutput\":[{\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"Cost vs Sales\",\"type\":\"report\",\"viewid\":[\"100003000000002258\"],\"username\":[\"abinesh.b@zohocorp.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"],\"emailname\":[\"Cost vs Sales Update\"],\"successmessage\":[\"The email has been successfully sent to abinesh.b@zohocorp.com with Cost vs Sales reports.\"]}}]}]},{\"currentFinalConsolidatedOutput\":[{\"REPORT\":[{\"type\":\"report\",\"viewid\":\"100003000000002258\",\"name\":\"Cost vs Sales\"}]}]},{\"previousFinalConsolidatedOutput\":[]}, {\n  \"user_query\": \"send this as mail\",\n  \"output\" : [ \n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"Cost vs Sales\", \"type\": \"report\", \"viewid\":[\"100004000000092555\"], \"exporttype\": [\"CSV\"],\"sharetype\":[\"mail\"], \"username\":[\"abinesh.b@zohocorp.com\"], \"usertype\":[\"person\"], \"emailname\":[\"Cost vs Sales Update\", \"successmessage\":[\"The email has been successfully sent to abinesh.b@zohocorp.com with Cost vs Sales reports.\"] } }\n    ]\n} ],\n[{\"previousQueryDetails\":[{\"query\":\"open sales chart\",\"extractedOutput\":[{\"action\":\"OPEN\",\"arguments\":{\"module\":\"CHART\",\"name\":\"sales\",\"type\":\"chart\"}}]},{\"query\":\"export the above chart as csv\",\"extractedOutput\":[{\"action\":\"EXPORT\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"Cost vs Sales\",\"type\":\"report\",\"viewid\":[\"100003000000002258\"],\"exporttype\":[\"CSV\"]}}]},{\"query\":\"share it to abinesh.b@zohocorp.com\",\"extractedOutput\":[{\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"Cost vs Sales\",\"type\":\"report\",\"viewid\":[\"100003000000002258\"],\"username\":[\"abinesh.b@zohocorp.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"],\"emailname\":[\"Cost vs Sales Update\"],\"successmessage\":[\"The email has been successfully sent to abinesh.b@zohocorp.com with Cost vs Sales reports.\"]}}]}]},{\"currentFinalConsolidatedOutput\":[{\"REPORT\":[{\"type\":\"report\",\"viewid\":\"100003000000002258\",\"name\":\"Cost vs Sales\"}]}]},{\"previousFinalConsolidatedOutput\":[]}, {\n  \"user_query\": \"schedule this at 10AM\",\n  \"output\" : [ \n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"Cost vs Sales\", \"type\": \"report\", \"viewid\":[\"100004000000092555\"], \"exporttype\": [\"CSV\"], \"username\":[\"abinesh.b@zohocorp.com\"], \"usertype\":[\"person\"], \"time\": [\"10:00\"],\"schedulername\":[\"Cost Vs Sales Analysis Schedule\"], \"successmessage\":[\"The email has been successfully scheduled with cost vs sales reports! It will be sent to abinesh.b@zohocorp.com at 10 PM.\"] } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open sales chart\",\n    \"extractedOutput\" : [ \n        { \"action\": \"OPEN\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales\", \"type\": \"chart\" } }\n      ]\n  } ]\n},\n  \"finalConsolidatedOutput\" : [ \n  { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"100004000000092555\"}, { \"type\" : \"Report\", \"viewid\" : \"1000040000000926556\"}, { \"type\" : \"Report\", \"viewid\" : \"10000400000067657\"} ]\n  } ]\n}, {\n  \"user_query\": \"schedule it as csv to manoj on Monday at 8PM\",\n  \"output\" : [ \n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"sales\", \"type\": \"chart\", \"time\": [\"20:00\"], \"viewid\": [\"100004000000092555\", \"1000040000000926556\", \"10000400000067657\"], \"exporttype\": [\"CSV\"], \"username\": [\"manoj\"], \"usertype\": [\"person\"], \"schedulername\": [\"Sales Performance Analysis Schedule\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to manoj on Monday at 8 PM.\"] } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales view\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"VIEW\",\"name\" : \"sales\",\"type\" : \"view\"}\n    } ]\n  } ]\n}, {\n  \"previousFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172326\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172146\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172154\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172126\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172118\"}]\n  } ]\n}, {\n  \"user_query\" : \"export pivot chart\",\n  \"output\" : [ \n    {\"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"type\" : \"report\"} }\n   ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales report\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : { \"module\" : \"REPORT\",\"name\" : \"sales\",\"type\" : \"report\"}\n    } ]\n  }, {\n    \"query\" : \"export pivot chart\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"type\" : \"pivot\"}\n    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"cost\",\n  \"output\" :[ \n    {\"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"name\" : \"cost\",\"type\" : \"pivot\"} }\n   ]\n} ][ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales report\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : { \"module\" : \"REPORT\",\"name\" : \"sales\",\"type\" : \"report\"}\n    } ]\n  }, {\n    \"query\" : \"export pivot chart\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"type\" : \"pivot\"}\n    } ]\n  }, {\n    \"query\" : \"cost\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"name\" : \"cost\",\"type\" : \"pivot\"}\n    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"Pivot | Report\",\"viewid\" : \"1000030000002452423\"} ]\n  } ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\" : \"share it to sandeep.kotra\",\n  \"output\" :[ {\n    \"action\" : \"SHARE\",\"arguments\" : {\"module\" : \"PIVOT\",\"name\" : \"cost\",\"type\" : \"pivot\", \"viewid\" : [\"1000030000002452423\"], \"username\": [\"sandeep.kotra\"], \"usertype\": [\"person\"],\"sharetype\": [\"analytics\"], \"emailname\": [\"Cost Update\"], \"successmessage\": [\"The email has been successfully sent to sandeep.kotra with cost pivot views.\"] }\n  } ]\n} ][ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export pivot charts\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"type\" : \"pivot\"}\n      } ]\n  }, {\n    \"query\" : \"sales\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"name\" : \"sales\",\"type\" : \"pivot\"}\n    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"Pivot | Report\",\"viewid\" : \"1000030000002452423\"} ]\n  } ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\" : \"send it as email to @manoj.gattam include insights also\",\n  \"output\" : [ \n    {\"action\" : \"SHARE\",\"arguments\" : {\"module\" : \"PIVOT\",\"name\" : \"sales\",\"type\" : \"pivot\",\"viewid\" : [\"1000030000002452423\"],\"username\" : [ \"manoj.gattam\" ],\"usertype\" : [ \"person\" ],\"sharetype\" : [ \"mail\" ], \"emailname\": [\"Sales Update\"], \"successmessage\": [\"The email has been successfully sent to manoj.gattam with sales reports.\"]}\n   ]\n} ][ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"list all table\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"LIST\",\"arguments\" : {\"module\" : \"TABLE\",\"type\" : \"table\"}\n    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"TABLE\" : [ {\"type\" : \"Table\",\"viewid\" : \"100003000000172326\"} ]\n  } ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\" : \"export the above table.\",\n  \"output\" : [ {\n    \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"TABLE\",\"name\" : \"name\",\"type\" : \"table\", \"viewid\" : [\"100003000000172326\"]}\n  } ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"schedule distribution of cost as excel to manoj.gattam@zohotest.com on august, sep and oct 5th\",\n    \"extractedOutput\" : [ {\n       {\"action\":\"SCHEDULE\",\"arguments\":{\"module\":\"MAIL\",\"name\":\"distribution of cost\",\"type\":\"view\", \"exporttype\": [\"XLS\"],\"username\":[\"manoj.gattam@zohotest.com\"],\"scheduletype\":[\"YEARLY\"],\"scheduleday\":[\"5\"],\"scheduleinterval\":[\"1\"],\"schedulemonth\":[\"7\",\"8\",\"9\"],\"successmessage\":[\"The email has been successfully scheduled! It will be sent to manoj.gattam@zohotest.com on 5th August, September, and October.\"],\"schedulername\":[\"Distribution of Cost Update Schedule\"]}}     } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"Pivot | Report\",\"viewid\" : \"1000030000002452423\"} ]\n  } \n  \"previousFinalConsolidatedOutput\" : [ ]\n}], {\n  \"user_query\" : \"10am\",\n  \"output\" : [ {\n       {\"action\":\"SCHEDULE\",\"arguments\":{\"module\":\"MAIL\",\"name\":\"distribution of cost\",\"type\":\"view\", \"viewid\" : [\"1000030000002452423\"],\"exporttype\": [\"XLS\"],\"username\":[\"manoj.gattam@zohotest.com\"],\"time\":[\"10:00\"],\"scheduletype\":[\"YEARLY\"],\"scheduleday\":[\"5\"],\"scheduleinterval\":[\"1\"],\"schedulemonth\":[\"7\",\"8\",\"9\"],\"successmessage\":[\"The email has been successfully scheduled! It will be sent to manoj.gattam@zohotest.com at 10 AM on 5th August, September, and October.\"],\"schedulername\":[\"Distribution of Cost Update Schedule\"]}}  } ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"list all reports\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"LIST\",\"arguments\" : {\"module\" : \"REPORT\",\"type\" : \"report\"}\n    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172326\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172154\"} ]\n  } ],\n  \"previousFinalConsolidatedOutput\" : [ {\n  } ]\n} ], {\n  \"user_query\" : \"export the first and third report.\",\n  \"output\" : [ \n    {\"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"REPORT\",\"name\":\"name\", \"type\" : \"report\", \"viewid\" : [\"100003000000172326\", \"100003000000172154\"]}\n  ]\n} ] [ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"share sales report to manoj.gattam@gmail.com\",\n    \"extractedOutput\" : [ {\n       {\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"sales\",\"type\":\"report\",\"username\":[\"manoj.gattam@gmail.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"], \"emailname\": [\"sales Update\"], \"successmessage\": [\"The email has been successfully sent to manoj.gattam@gmail.com with sales reports.\"]}}    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172326\"}]\n  } ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\" : \"schedule this as pdf everyday\",\n  \"output\" : [ {\n    \"action\" : \"SCHEDULE\",\n    \"arguments\" : {\"module\" : \"MAIL\",\"name\" : \"sales\",\"type\" : \"report\",\"viewid\" : [\"100003000000172326\"], \"exporttype\": [\"PDF\"],\"username\" : [ \"manoj.gattam@gmail.com\" ],\"scheduletype\" : [ \"DAILY\" ],\"successmessage\" : [ \"The email has been successfully scheduled! It will be sent to manoj.gattam@zohocorp.com every day.\" ],\"schedulername\":[\"Sales Performance Analysis Schedule\"]}   } ]\n} ][ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open sales report\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"OPEN\",\"arguments\" : {\"module\" : \"REPORT\",\"name\" : \"sales\",\"type\" : \"report\"}\n    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000296003\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"1000030000002645344\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"10000300000236743\"} ]\n  } ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\" : \"also send this as mail\",\n  \"output\" : [ {\n    \"action\" : \"SHARE\",\"arguments\" : {\"module\" : \"REPORT\",\"name\" : \"sales\",\"type\" : \"report\",\"viewid\" : [ \"100003000000296003\" ],\"sharetype\" : [ \"mail\" ]}\n  } ]\n} ],[{\n    \"previousQueryDetails\" : [ {\n      \"query\" : \"export cost and sales view\",\n      \"extractedOutput\" : [ \n      { \"action\" : \"EXPORT\",\"arguments\" : { \"module\" : \"VIEW\",\"name\" : \"cost\",\"type\" : \"view\"}},\n      { \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"VIEW\",\"name\" : \"sales\",\"type\" : \"view\"}}\n    ]},\n     {\n      \"query\" : \"export first three reports\",\n      \"extractedOutput\" : [ {\n        \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"REPORT\",\"viewid\" : [ \"100003000000253211\", \"10000300005345325\", \"10000300000124124\" ], \"type\" : \"report\"}\n      } ]\n    } ]\n  }, \n  {\n    \"currentFinalConsolidatedOutput\" : [\n       {\"REPORT\" : [ {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000253211\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"10000300005345325\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"10000300000124124\"} ]}\n    ],\n    \"previousFinalConsolidatedOutput\" : [\n       {\"DASHBOARD\" : [ {\"type\" : \"Dashboard\",\"viewid\" : \"1000030000003124534\"}, {\"type\" : \"Dashboard\",\"viewid\" : \"100003000001241435\"}, {\"type\" : \"Dashboard\",\"viewid\" : \"10000300001244523\"} ]}\n    ]\n  }, \n  {\n    \"user_query\" : \"also share them to dinesh.ravindran@gmail.com\",\n    \"output\":[ \n        {\"action\" : \"SHARE\",\"arguments\" : {\"module\" : \"REPORT\",\"name\":\"cost\", \"viewid\" : [ \"100003000000253211\", \"10000300005345325\", \"10000300000124124\" ],\"type\" : \"report\",\"username\" : [ \"dinesh.ravindran@gmail.com\" ],\"usertype\" : [ \"person\" ],\"sharetype\" : [ \"analytics\" ], \"emailname\": [\" Cost and Sales Update\"], \"successmessage\": [\"The email has been successfully sent to dinesh.ravindran@gmail.com with cost and sales reports.\"]}}\n      ]\n  }] ]}"}

za.zia.action.slideshowtitle.extraction="You are a title generation assistant.\nYour job is to generate a short, clear, professional business title based on the user’s query.\n\n### Output format:\n{\n  \"criterias\": [\n    {\n      \"key\": \"title\",\n      \"value\": [\"your generated title here\"]\n    }\n  ]\n}\n\n### Guidelines:\n- Only use letters, digits, spaces, and non-Basic Latin characters.\n- Avoid punctuation: :, ;, ,, ?, !, etc.\n- Use consistent title casing (capitalize important words).\n- The title should directly match the user’s intent.\n- Must match this regex: ([0-9a-zA-Z\\P{InBasicLatin}\\s]+$)\n- STRICTLY FOLLOW the output format above. Return only the JSON.\n\n### Few-shot examples:\nUser: create a slideshow for monthly sales performance.\nResponse:\n{\"criterias\": [{\"key\": \"title\", \"value\": [\"Monthly Sales Performance Report\"]}]}\n\nUser: send an email scheduler for weekly sales charts.\nResponse:\n{\"criterias\": [{\"key\": \"title\", \"value\": [\"Weekly Sales Charts Update\"]}]}\n\nUser: create a slideshow for marketing campaign results.\nResponse:\n{\"criterias\": [{\"key\": \"title\", \"value\": [\"Marketing Campaign Results Summary\"]}]}\n\nUser: send an email scheduler for quarterly financial highlights.\nResponse:\n{\"criterias\": [{\"key\": \"title\", \"value\": [\"Quarterly Financial Highlights\"]}]}\n\nUser: create a slideshow for customer satisfaction survey insights.\nResponse:\n{\"criterias\": [{\"key\": \"title\", \"value\": [\"Customer Satisfaction Insights\"]}]}\n\nUser: send an email scheduler for daily inventory updates.\nResponse:\n{\"criterias\": [{\"key\": \"title\", \"value\": [\"Daily Inventory Status\"]}]}\n\nUser: create a slideshow for annual revenue growth analysis.\nResponse:\n{\"criterias\": [{\"key\": \"title\", \"value\": [\"Annual Revenue Growth Report\"]}]}\n\nUser: send an email scheduler for project milestone updates.\nResponse:\n{\"criterias\": [{\"key\": \"title\", \"value\": [\"Weekly Project Milestones\"]}]}\n\nUser: create a slideshow for product launch performance metrics.\nResponse:\n{\"criterias\": [{\"key\": \"title\", \"value\": [\"Product Launch Performance Review\"]}]}\n\nUser: send an email scheduler for team productivity dashboards.\nResponse:\n{\"criterias\": [{\"key\": \"title\", \"value\": [\"Monthly Team Productivity Report\"]}]}"

za.askzia.action.qwen.prompt.for.zia.turbo={"actions":"{\"SCHEDULE\":\"If the user wants to schedule any chart, report, view, dashboard, table, or query table, then STRICTLY interpret the action as 'SCHEDULE'.\",\"SHARE\":\"If the user wants to 'share' or 'send' any chart, report, view, dashboard, table, or query table, then strictly interpret the action as 'SHARE'.\",\"LIST\":\"If the user wants to 'list' or 'get' any chart, report, view, dashboard, table, or query table, then strictly interpret the action as 'LIST', regardless of whether the original verb used was 'list' or 'get'.STRICTLY always normalize these verbs to the single action: 'LIST'\",\"EXPORT\":\"If the user wants to export any chart, report, view, dashboard, table, or query table, then strictly interpret the action as 'EXPORT'.\",\"OPEN\":\"If the user wants to 'open', 'show', 'fetch', or 'display' any chart, report, view, dashboard, table, or query table, then strictly interpret the action as 'OPEN', regardless of whether the original verb used was 'show', 'display', or 'fetch'.STRICTLY always normalize these verbs to the single action: 'OPEN'.\"}","modules":"{\"DASHBOARD\":\"always use 'dashboard' as the type for any request that involves a dashboard.\",\"CHART\":\"always use 'chart' as the type for any request that involves charts .\",\"REPORT\":\"always use 'report' as the module type for any request involving reports.\",\"SUMMARYVIEW\":\"always use 'summaryview' as the type for any request that involves a summary view or summary charts.\",\"TABULARVIEW\":\"always use 'tabularview' as the type for any request that involves a tabular view or tabular charts.\",\"VIEW\":\"always use 'view' as the type for any request that has no 'type' param to be extracted. Assign 'view' by default.\",\"PIVOT\":\"always use 'pivot' as the type for any request that involves a pivot view or pivot charts.\",\"QUERYTABLE\":\"always use querytable as the type for any request that involves a querytable.\",\"TABLE\":\"always use table as the type for any request that involves a table.\"}","Parameters to Extract:":"{\"name\":\"Extract the exact title/name of the report, view, pivot, summaryview, scheduled mail name, table, query table, dashboard.Do not include any additional qualifiers like 'charts', 'report', 'view', or 'dashboard'.STRICTLY If there is NO name is explicitly not mentioned and if you have any resolving confusion, leave it as null.\",\"username\":\"Extract the username or email only if it's explicitly mentioned, regardless of the action.\",\"time\":\"This task is specifically for scheduling an email or creating a scheduler skill. All time values must be extracted and represented strictly in 24-hour format (HH:MM). STRICTLY If no specific time is mentioned in the query, do not assume or assign a default time like 00:00 \\u2014 EXCLUDE this param entirely from the output.\",\"type\":\"Extract the exact type (report/table/dashboard/querytable/view/summaryview/pivot/tabularview) only if explicitly stated, otherwise return null for any ambiguity or absence of explicit mention.STRICTLY tag 'chart' and 'view' as 'report'.\",\"exporttype\":\"extract the format to export (if explicitly stated). only return values like 'CSV', 'PDF', 'IMG' etc. strictly treat 'IMAGE' as an export type, and convert it to 'IMG'.\",\"scheduletype\":\"Decide the schedule type for a mail or scheduler. Return only one of these values as a string: \\\"DAILY\\\", \\\"WEEKLY\\\", \\\"MONTHLY\\\", or \\\"YEARLY\\\". Use \\\"DAILY\\\" for daily patterns, \\\"WEEKLY\\\" for weekly patterns or weekdays, \\\"MONTHLY\\\" for anything happening every month or every few months, and \\\"YEARLY\\\" for anything related to specific months or yearly events.\",\"scheduleinterval\":\"This indicates the frequency of the scheduled task. STRICTLY extract it as an integer ranging from 1 to 4, where 1 indicates the first week of the month, 2 indicates the second week of the month, 3 indicates the third week of the month, and 4 indicates the fourth week of the month. For example, \\u201cevery Monday\\u201d would be 1, \\u201cevery second Monday of the month\\u201d would be 2, and \\u201cevery third Monday of the month\\u201d would be 3.\",\"scheduleday\":\"This is used for scheduling an email or creating a scheduler skill. You must strictly extract all values as strings representing integers. If the query refers to a weekly schedule, the days of the week must be mapped using the following exact mapping: Sunday = \\\"1\\\", Monday = \\\"2\\\", TUESDAY = \\\"3\\\", Wednesday = \\\"4\\\", Thursday = \\\"5\\\", Friday = \\\"6\\\", and Saturday = \\\"7\\\". STRICTLY IF IT \\\"TUESDAY\\\",MAP IT TO \\\"3\\\". Do not return day names like \\\"Tuesday\\\" \\u2014 only return the corresponding mapped string values exactly as specified.If multiple weekdays are mentioned, return them as a string array of these mapped values (e.g., [\\\"2\\\", \\\"4\\\", \\\"6\\\"]). If the query refers to a yearly schedule, extract the day of the month as a string integer between \\\"1\\\" and \\\"31\\\".If multiple values are mentioned, return them as a comma-separated array of string integers.\",\"schedulemonth\":\"This is specifically used for scheduling a mail or creating a scheduler skill. If the query refers to a yearly schedule, you must STRICTLY extract the month as an integer between 0 and 11, where the value \\\"January\\\"=\\\"0\\\" and \\\"December\\\"=\\\"11\\\". The mapping must follow this exact order: \\\"January\\\"=\\\"0\\\", \\\"February\\\"=\\\"1\\\", \\\"March\\\"=\\\"2\\\", \\\"April\\\"=\\\"3\\\", \\\"May\\\"=\\\"4\\\", \\\"June\\\"=\\\"5\\\", \\\"July\\\"=\\\"6\\\", \\\"August\\\"=\\\"7\\\", \\\"September\\\"=\\\"8\\\", \\\"October\\\"=\\\"9\\\", \\\"November\\\"=\\\"10\\\", and \\\"December\\\"=\\\"11\\\". Do not extract or return any values outside this range.\",\"schedulername\":\"STRICTLY AUTOMATICALLY GENERATE THE SCHEDULERNAME USING THE NAME, USERNAME, AND TIME FIELDS.\",\"successmessage\":\"STRICTLY Automatically generate the successmessage using the name, username fields\",\"emailname\":\"STRICTLY Automatically generate the email name using the name fields\",\"usertype\":\"Describes type of user in user information to share views with.It contains one of the following values person, channel, group.\",\"sharetype\":\"Describes medium through which views can be shared.It cotains one of the following values cliq, slack etc. STRICTLY include medium if it is present explicitly in the input.\",\"viewid\":\"STRICTLY Extract the exact 'viewid' of the report, table, query table, dashboard ONLY IF the previous function response 'type' EXACTLY matches the current request 'type'. If the types DO NOT match, DO NOT include 'viewid' at all in the function arguments. If there is ambiguity, return null and exclude 'viewid' without making assumptions.\"}","Guidelines":"Only include parameters that are explicitly mentioned by the user.\\n- **STRICTLY IF 'MODULE' AND 'TYPE' ARE NOT FOUND IN THE QUERY, SET 'MODULE' TO 'VIEW' AND 'TYPE' TO 'VIEW'.**\\n- **STRICTLY** GENERATE OUTPUT IN PROPER GIVEN FORMAT - **\"{\"output\": \"ans\"}\"** WITHOUT ANY SYNTAX ISSUES AND 'ans' MUST BE A JSON ARRAY.\\n- **STRICTLY ALWAYS THE RESPONSE IS A JSON ARRAY, WRAP IT IN AN OBJECT WITH 'output' AS THE KEY AND THE JSON ARRAY AS THE VALUE.**\\n- **STRICTLY RETURN OUTPUT FOR ONLY ONE ACTION AND ONE MODULE. IF MULTIPLE CANDIDATES ARE PRESENT, SELECT ONLY THE MOST RECENTLY USED ACTION AND MODULE FROM THE QUERY CONTEXT. DO NOT RETURN MULTIPLE ACTIONS OR MODULES IN THE OUTPUT. EVEN IF FINAL CONSOLIDATED OUTPUT CONTAINS MULTIPLE MATCHES, CHOOSE ONLY THE ONE MOST CLOSELY AND RECENTLY REFERENCED. IN CASE OF AMBIGUITY, PRIORITIZE THE MOST RECENT SEMANTICALLY RELEVANT ACTION AND MODULE FROM THE USER QUERIES.- **STRICTLY EXTRACT THE USERNAME ONLY IF THE EMAIL IS PRESENT IN THE MOST RECENT SEMANTICALLY RELEVANT QUERY.** IF ANY OTHER ACTION SUCH AS `OPEN`, `EXPORT`, OR ANY NON-SHARE/NON-SCHEDULE ACTION INTERRUPTS THE CONTEXT, DO NOT CARRY FORWARD OR EXTRACT THE USERNAME FROM OLDER QUERIES. CONTEXTUAL CONTINUITY MUST BE MAINTAINED ONLY WHEN THE EMAIL IS STILL SEMANTICALLY RELEVANT TO THE CURRENT QUERY.\\n  - **STRICTLY FOR THE `SCHEDULE` ACTION, EXTRACT THE `USERNAME` ONLY IF AN EMAIL ADDRESS IS PRESENT IN THE MOST RECENT SEMANTICALLY RELEVANT QUERY.** IF ANY OTHER ACTION SUCH AS `OPEN`, `EXPORT`, OR `SHOW` OCCURS AFTER THE EMAIL, CONSIDER THE CONTEXT INTERRUPTED AND DO NOT EXTRACT THE `USERNAME` FROM OLDER QUERIES. `SCHEDULE` MUST ONLY USE THE EMAIL IF IT DIRECTLY FOLLOWS OR LOGICALLY DEPENDS ON THAT EMAIL WITHOUT ANY INTERVENING ACTION. IF THERE IS NO EMAIL IN THE CURRENT OR DIRECTLY LINKED CONTEXT, DO NOT RETURN THE `USERNAME` FIELD.\\n- **STRICTLY AND VERY IMPORTANT RETURN OUTPUT ONLY FOR THE MOST RECENT USER QUERY AND ITS CORRESPONDING ACTION AND MODULE. DO NOT INCLUDE RESPONSES FOR ANY PREVIOUS ACTIONS OR QUERIES.ONLY A SINGLE ACTION AND SINGLE MODULE PER INVOCATION — IF MULTIPLE ACTIONS AND MODULES ARE PRESENT IN CONTEXT OR HISTORY, RESPOND ONLY TO THE MOST RECENT ACTION AND MODULE VALID ONE.\\n- **IMPORTANT: DO NOT COPY OR REUSE ANY CONTENT FROM THE FEW-SHOT EXAMPLES. THE EXAMPLES ARE FOR STRUCTURE REFERENCE ONLY AND CONTAIN RANDOM DATA. THE MODEL MUST GENERATE OUTPUT BASED ONLY ON THE ACTUAL USER INPUT .**\\n- **STRICTLY JUST EXTRACT THE REQUIRED VIEWIDS FROM THE 'finalConsolidatedOutput' and put it in 'VIEWID' JSONARRAY AND COPY 'NAME' PARAMETER FROM THE 'previousextractedoutput'.**\\n- **STRICTLY ENSURE THAT ONLY IF THE 'VIEWID' KEY IS PRESENT AND THE 'NAME' PARAMETER IS MISSING, THE 'NAME' PARAMETER MUST BE ASSIGNED THE VALUE 'NAME'. THE 'NAME' KEY MUST ALWAYS BE PRESENT IN THE OUTPUT EXTRACTION WHENEVER 'VIEWID' IS PRESENT.**\\n- **STRICTLY ENSURE TO HAVE A CLEAR DISTINCTION IS MAINTAINED BETWEEN 'REPORT', 'CHART', 'PIVOT', AND 'SUMMARYVIEW'.**\\n- STRICTLY 'viewid' MUST be included ONLY if the previous function response’s 'type' EXACTLY matches the current query's requested 'type'.\\n","Guardrails":"STRICTLY only the actions and modules explicitly provided in the prompt and few-shot examples are supported. If any other action or module is present, respond with: {\"error\":\"This operation is not supported.\"}\\n\\nSTRICTLY NO actions on slideshow are not at all supported. STRICTLY ONLY CREATE action on dashboard module is not supported. All others are supported.\\n\\nUnsupported examples:\\n- Save the above chart — Save is not supported.\\n- Open Zia settings — Settings module not supported.\\n- create a slideshow using sales metric — slideshow module not supported.\\n- Prepare a presentation of sales distribution across customers and regions . — slideshow module not supported.\\n- change the theme of profit analysis slideshow — slideshow module not supported.\\n- create a dashboard using cost metric — dashboard creation is not supported.\\n- Add Manoj as workspace admin — Admin actions not supported.\\n- Share to MS Teams, Slack, Cliq — External app sharing is not supported.\\n\\nSTRICTLY unsupported action-module combinations:\\n- Create/Add/Rename — Table, Column, Alert, User, User group\\n- Open — Zia, Org, Workspace Settings, Model Diagram, Data Model\\n- Share to — Slack, Cliq, Support, Publish\\n- Delete/Revoke/Edit — Schedulers, Permissions, Lookups, Tables, Reports, Alerts, Users\\n- Schedule — Sync\\n- List — Email Schedules, Users, Alerts, Columns, Shared views, Variables, Sync schedules\\n","Few Shot Examples":"{\n  \"FEW SHOT Examples\": [\n    {\n      \"user_query\": \"Export\",\n      \"output\": [\n        { \"action\": \"EXPORT\", \"arguments\": {\"module\": \"VIEW\", \"type\": \"view\"} }\n      ]\n    },\n    {\n      \"user_query\": \"export all summary views\",\n      \"output\": [\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"type\": \"summaryview\" } }\n      ]\n    },\n    {\n      \"user_query\": \"export all tabular views\",\n      \"output\": [\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"TABULARVIEW\" \"type\": \"tabularview\" } }\n      ]\n    },\n    {\n      \"user_query\": \"export sales\",\n      \"output\": [\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"sales\", \"type\": \"view\" } },\n      ]\n    },\n    {\n      \"user_query\": \"export customer view\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"customer\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"export sales chart as image\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales\", \"type\": \"chart\", \"exporttype\": [\"IMG\"] } } ]\n    },\n    {\n      \"user_query\": \"export Moving Average of Cost Across Months in 2010\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"Moving Average of Cost Across Months in 2010\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"export Year to Date ( YTD ) Sales by each Year\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"Year to Date ( YTD ) Sales by each Year\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"export region summary view as jpg\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"region\", \"type\": \"summaryview\", \"exporttype\": [\"IMG\"] } } ]\n    },\n    {\n      \"user_query\": \"export sales pivot\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales\", \"type\": \"pivot\" } } ]\n    },\n    {\n      \"user_query\": \"export sales by region across years\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"sales by region across years\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"export product querytable as xls\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"QUERYTABLE\", \"name\": \"product\", \"type\": \"querytable\", \"exporttype\": [\"XLS\"] } } ]\n    },\n    {\n      \"user_query\": \"export feedback tabular view as html\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"TABULARVIEW\", \"name\": \"feedback\", \"type\": \"tabularview\", \"exporttype\": [\"HTML\"] } } ]\n    },\n    {\n      \"user_query\": \"export technology table\",\n      \"output\": [ { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"TABLE\", \"name\": \"technology\", \"type\": \"table\" } } ]\n    }\n    {\n      \"user_query\": \"export profit across years pivot charts as excel and image\",\n      \"output\": [\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"profit across years\", \"type\": \"pivot\", \"exporttype\": [\"XLS\", \"IMG\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"Export sales pivot tabular view and profit summary view as HTML\",\n      \"output\": [\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"TABULARVIEW\", \"name\": \"sales\", \"type\": \"tabularview\", \"exporttype\": [\"HTML\"] } },\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"profit\", \"type\": \"summaryview\", \"exporttype\": [\"HTML\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share product performance view\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"product performance\", \"type\": \"view\", \"sharetype\": [\"analytics\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share revenue report to @jane.doe\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"revenue\", \"type\": \"report\", \"username\": [\"jane.doe\"], \"usertype\": [\"person\"], \"sharetype\": [\"analytics\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"email marketing summary view as html to sam\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"marketing\", \"type\": \"summaryview\", \"exporttype\": [\"HTML\"], \"username\": [\"sam\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Marketing Summary\"], \"successmessage\": [\"The email has been successfully sent to sam with marketing summary views as HTML.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share sales trends pivot as image with @group-marketing\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales trends\", \"type\": \"pivot\", \"exporttype\": [\"IMG\"], \"username\": [\"group-marketing\"], \"usertype\": [\"group\"], \"sharetype\": [\"analytics\"], \"emailname\": [\"Sales Trends Report\"], \"successmessage\": [\"The email has been successfully sent to group-marketing with sales trends pivot as IMAGE.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send inventory chart to david without insights\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"CHART\", \"name\": \"inventory\", \"type\": \"chart\", \"username\": [\"david\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Inventory Chart\"], \"successmessage\": [\"The email has been successfully sent to david with inventory charts.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send sales report to sandeep.kotra@zohocorp.com\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"sales\", \"type\": \"report\", \"username\": [\"sandeep.kotra@zohocorp.com\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Sales Chart Update\"], \"successmessage\": [\"The email has been successfully sent to sandeep.kotra@zohocorp.com with sales charts.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send profit to manoj.gattam\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"profit\", \"type\": \"view\", \"username\": [\"manoj.gattam\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Profit Chart Update\"], \"successmessage\": [\"The email has been successfully sent to manoj.gattam with profit views.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share revenue dashboard to finance group and disable commenting\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"revenue\", \"type\": \"dashboard\", \"username\": [\"finance\"], \"usertype\": [\"group\"], \"sharetype\": [\"analytics\"], \"emailname\": [\"Revenue Dashboard Report\"], \"successmessage\": [\"The email has been successfully sent to finance with revenue dashboards.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"email customer orders table to diana\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"TABLE\", \"name\": \"customer orders\", \"type\": \"table\", \"username\": [\"diana\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Customer Orders Table\"], \"successmessage\": [\"The email has been successfully sent to diana with customer orders table.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send weekly performance querytable to @alex as html\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"QUERYTABLE\", \"name\": \"weekly performance\", \"type\": \"querytable\", \"exporttype\": [\"HTML\"], \"username\": [\"alex\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Weekly Performance\"], \"successmessage\": [\"The email has been successfully sent to alex with weekly performance query table as HTML.\"] } }\n      ]\n    }\n    {\n      \"user_query\": \"share product performance view\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"product performance\", \"type\": \"view\", \"sharetype\": [\"analytics\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share revenue report to @jane.doe\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"revenue\", \"type\": \"report\", \"username\": [\"jane.doe\"], \"usertype\": [\"person\"], \"sharetype\": [\"analytics\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"email marketing summary view as html to sam\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"marketing\", \"type\": \"summaryview\", \"exporttype\": [\"HTML\"], \"username\": [\"sam\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Marketing Summary\"], \"successmessage\": [\"The email has been successfully sent to sam with marketing summary views as HTML.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share sales trends pivot as image with @group-marketing\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales trends\", \"type\": \"pivot\", \"exporttype\": [\"IMG\"], \"username\": [\"group-marketing\"], \"usertype\": [\"group\"], \"sharetype\": [\"analytics\"], \"emailname\": [\"Sales Trends Report\"], \"successmessage\": [\"The email has been successfully sent to group-marketing with sales trends pivot as IMAGE.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send inventory chart to david without insights\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"CHART\", \"name\": \"inventory\", \"type\": \"chart\", \"username\": [\"david\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Inventory Chart\"], \"successmessage\": [\"The email has been successfully sent to david with inventory charts.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"share revenue dashboard to finance group and disable commenting\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"revenue\", \"type\": \"dashboard\", \"username\": [\"finance\"], \"usertype\": [\"group\"], \"sharetype\": [\"analytics\"], \"emailname\": [\"Revenue Dashboard Report\"], \"successmessage\": [\"The email has been successfully sent to finance with revenue dashboards.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"email customer orders table to diana\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"TABLE\", \"name\": \"customer orders\", \"type\": \"table\", \"username\": [\"diana\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Customer Orders Table\"], \"successmessage\": [\"The email has been successfully sent to diana with customer orders table.\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send weekly performance querytable to @alex as html\",\n      \"output\": [\n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"QUERYTABLE\", \"name\": \"weekly performance\", \"type\": \"querytable\", \"exporttype\": [\"HTML\"], \"username\": [\"alex\"], \"usertype\": [\"person\"], \"sharetype\": [\"mail\"], \"emailname\": [\"Weekly Performance\"], \"successmessage\": [\"The email has been successfully sent to alex with weekly performance query table as HTML.\"] } }\n      ]\n    }\n    {\n      \"user_query\": \"list sales\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"sales\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"list all customer charts\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"CHART\", \"name\": \"customer\", \"type\": \"chart\" } } ]\n    },\n    {\n      \"user_query\": \"list all tabular views for revenue\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"TABULARVIEW\", \"name\": \"revenue\", \"type\": \"tabularview\" } } ]\n    },\n    {\n      \"user_query\": \"list all query tables\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"QUERYTABLE\", \"type\": \"querytable\" } } ]\n    },\n    {\n      \"user_query\": \"list sales reports\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"sales\", \"type\": \"report\" } } ]\n    },\n    {\n      \"user_query\": \"list performance dashboards\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"performance\", \"type\": \"dashboard\" } } ]\n    },\n    {\n      \"user_query\": \"list product tables\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"TABLE\", \"name\": \"product\", \"type\": \"table\" } } ]\n    },\n    {\n      \"user_query\": \"get all summary views for quarterly revenue\",\n      \"output\": [ { \"action\": \"LIST\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"quarterly revenue\", \"type\": \"summaryview\" } } ]\n    },\n    {\n      \"user_query\": \"open regional performance view\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"regional performance\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"show sales distribution chart\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales distribution\", \"type\": \"chart\" } } ]\n    },\n    {\n      \"user_query\": \"open cost by years by region\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"cost by years by region\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"open sales summary view\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"sales\", \"type\": \"summaryview\" } } ]\n    },\n    {\n      \"user_query\": \"open quarterly profit pivot\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"quarterly profit\", \"type\": \"pivot\" } } ]\n    },\n    {\n      \"user_query\": \"show me the dashboards\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"DASHBOARD\", \"type\": \"dashboard\" } } ]\n    },\n    {\n      \"user_query\": \"show customer overview summaryview\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"customer overview\", \"type\": \"summaryview\" } } ]\n    },\n    {\n      \"user_query\": \"open Sales and Cost for Years and Product Category by Region\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"Sales and Cost for Years and Product Category by Region\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"open employee list table\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"TABLE\", \"name\": \"employee list\", \"type\": \"table\" } } ]\n    },\n    {\n      \"user_query\": \"show budget vs actual tabular view\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"TABULARVIEW\", \"name\": \"budget vs actual\", \"type\": \"tabularview\" } } ]\n    },\n    {\n      \"user_query\": \"open product category wise sales\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"VIEW\", \"name\": \"open product category wise sales\", \"type\": \"view\" } } ]\n    },\n    {\n      \"user_query\": \"open marketing insights dashboard\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"marketing insights\", \"type\": \"dashboard\" } } ]\n    },\n    {\n      \"user_query\": \"show support queries querytable\",\n      \"output\": [ { \"action\": \"OPEN\", \"arguments\": { \"module\": \"QUERYTABLE\", \"name\": \"support queries\", \"type\": \"querytable\" } } ]\n    },\n    {\n      \"user_query\": \"schedule\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\",  \"type\": \"view\"} }]\n    },\n    {\n      \"user_query\": \"schedule sales report as pdf every monday at 9 am to john.doe\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"sales report\", \"type\": \"view\", \"exporttype\": [\"PDF\"], \"username\": [\"john.doe\"], \"time\": [\"09:00\"], \"scheduletype\": [\"WEEKLY\"], \"scheduleday\": [\"2\"], \"scheduleinterval\": [\"1\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to john.doe every Monday at 9 AM.\"], \"schedulername\": [\"Weekly Sales Report\"] } }]\n    },\n    {\n      \"user_query\": \"create a schedule to send profit analysis chart as image to alice on every thursday at 5 PM\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"profit analysis\", \"type\": \"chart\", \"exporttype\": [\"IMG\"], \"username\": [\"alice\"], \"time\": [\"17:00\"], \"scheduletype\": [\"WEEKLY\"], \"scheduleday\": [\"5\"], \"scheduleinterval\": [\"1\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to alice every Thursday at 5 PM.\"], \"schedulername\": [\"Profit Chart Weekly\"] } } ]\n    },\n    {\n      \"user_query\": \"schedule monthly summary of revenue trends as excel to mark at 8 AM\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"revenue trends\", \"type\": \"summaryview\", \"exporttype\": [\"XLS\"], \"username\": [\"mark\"], \"time\": [\"08:00\"], \"scheduletype\": [\"MONTHLY\"], \"scheduleday\": [\"1\"], \"scheduleinterval\": [\"1\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to mark on the 1st of every month at 8 AM.\"], \"schedulername\": [\"Monthly Revenue Summary\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"schedule cost insights pivot as csv every 3rd month to emma at 10 AM\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"cost insights\", \"type\": \"pivot\", \"exporttype\": [\"CSV\"], \"username\": [\"emma\"], \"time\": [\"10:00\"], \"scheduletype\": [\"MONTHLY\"], \"scheduleday\": [\"1\"], \"scheduleinterval\": [\"3\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to emma on the 1st day of every 3rd month at 10 AM.\"], \"schedulername\": [\"Quarterly Cost Pivot\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"send daily update of inventory table as png to raj at 6 AM\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"inventory\", \"type\": \"table\", \"exporttype\": [\"IMG\"], \"username\": [\"raj\"], \"time\": [\"06:00\"], \"scheduletype\": [\"DAILY\"], \"scheduleinterval\": [\"1\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to raj daily at 6 AM.\"], \"schedulername\": [\"Inventory Daily Update\"] } }\n      ]\n    },\n    {\n      \"user_query\": \"create annual schedule for product overview dashboard as image to sarah on January 15\",\n      \"output\": [\n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"product overview\", \"type\": \"dashboard\", \"exporttype\": [\"IMG\"], \"username\": [\"sarah\"], \"scheduletype\": [\"YEARLY\"], \"scheduleday\": [\"15\"], \"schedulemonth\": [\"0\"], \"scheduleinterval\": [\"1\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to sarah on 15th January of every year.\"], \"schedulername\": [\"Annual Product Dashboard\"] } }\n      ]\n    },\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales charts\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"EXPORT\", \"arguments\" : { \"module\" : \"CHART\", \"name\" : \"sales\", \"type\" : \"chart\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"analysisview | Report\", \"viewid\" : \"100003000008735857\"}, { \"type\" : \"Report\",\"viewid\" : \"1000030000002345222\"}, {\"type\" : \"pivot | Report\",\"viewid\" : \"100003000000234455\"}, {\"type\" : \"summaryview | Report\",\"viewid\" : \"1000030000001244553\"}, {\"type\" : \"tableview | Report\",\"viewid\" : \"1000030000035436436\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n    { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"1000030000005346634\"}, { \"type\" : \"Dashboard\", \"viewid\" : \"1000030000008687792\"} ] }\n  ]\n}, {\n  \"user_query\" : \"as pdf | as csv | as xls.\",\n  \"output\" : [ \n      {  \"action\" : \"EXPORT\", \"arguments\" : { \"module\" : \"CHART\", \"name\" : \"sales\",  \"type\" : \"chart\", \"viewid\":[\"100003000008735857\", \"1000030000002345222\", \"100003000000234455\", \"1000030000001244553\", \"1000030000035436436\"], \"exporttype\": [\"PDF\"] } } \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales analysis dashboards as xls.\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"EXPORT\", \"arguments\" : { \"module\" : \"DASHBOARD\", \"name\" : \"sales analysis\", \"type\" : \"dashboard\", \"exporttype\": [\"XLS\"] } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"1000030000005346634\"}, { \"type\" : \"Dashboard\", \"viewid\" : \"1000030000005346237\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"export top 2 dashboards.| export the above dashboards. \",\n  \"output\" : [ \n      {  \"action\" : \"EXPORT\", \"arguments\" : { \"module\" : \"DASHBOARD\", \"name\" : \"sales analysis\", \"type\" : \"dashboard\", \"viewid\":[\"1000030000005346634\", \"1000030000005346237\"], \"exporttype\": [\"XLS\"] } } \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open profit tabular view.\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"OPEN\", \"arguments\" : { \"module\" : \"TABULARVIEW\", \"name\" : \"profit\", \"type\" : \"tabularview\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"tabulerview | Report\", \"viewid\" : \"1000030000005346634\", \"name\" : \"profit\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"share this to sandeep.kotra as jpg.\",\n  \"output\" : [ \n      {  \"action\" : \"SHARE\", \"arguments\" : { \"module\" : \"TABULARVIEW\", \"name\" : \"profit\", \"type\" : \"tabularview\", \"exporttype\": [\"IMG\"],  \"viewid\":[\"1000030000005346634\"], \"username\": [\"sandeep.kotra\"], \"usertype\": [\"person\"],\"sharetype\": [\"analytics\"], \"emailname\": [\"Profit Updated Report\"], \"successmessage\": [\"The email has been successfully sent to sandeep.kotra with profit tabular views as CSV.\"] } \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open distribution of cost.\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"OPEN\", \"arguments\" : { \"module\" : \"VIEW\", \"name\" : \"distribution of cost\", \"type\" : \"view\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"1000030000005346634\", \"name\" : \"distribution of cost\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"email this view as csv and send to vinoglora.i+test1@zohotest.com\",\n  \"output\" : [ \n      {\"action\":\"SHARE\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"distribution of cost\",\"type\":\"view\", \"exporttype\": [\"CSV\"], \"viewid\":[\"1000030000005346634\"],\"username\":[\"vinoglora.i+test1@zohotest.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"mail\"], \"emailname\": [\"Distribution of Cost Update\"], \"successmessage\": [\"The email has been successfully sent to vinoglora.i+test1@zohotest.com with distribution of cost reports.\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open distribution of cost view\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"OPEN\", \"arguments\" : { \"module\" : \"VIEW\", \"name\" : \"distribution of cost\", \"type\" : \"view\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"1000030000005346634\", \"name\" : \"distribution of cost\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"share this view to vinoglora.i+free@zohotest.com.\",\n  \"output\" : [ \n      {\"action\":\"SHARE\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"distribution of cost\",\"type\":\"view\", \"viewid\":[\"1000030000005346634\"],\"username\":[\"vinoglora.i+free@zohotest.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"], \"emailname\": [\"Distribution of Cost Update\"], \"successmessage\": [\"The email has been successfully sent to vinoglora.i+free@zohotest.com with distribution of cost views.\"] }}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open distribution of cost view\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"OPEN\", \"arguments\" : { \"module\" : \"VIEW\", \"name\" : \"distribution of cost\", \"type\" : \"view\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"1000030000005346634\", \"name\" : \"distribution of cost\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"share this view to vinoglora.i+free@zohotest.com.\",\n  \"output\" : [ \n      {\"action\":\"SHARE\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"distribution of cost\",\"type\":\"view\", \"viewid\":[\"1000030000005346634\"],\"username\":[\"vinoglora.i+free@zohotest.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"], \"emailname\": [\"Distribution of Cost Update\"], \"successmessage\": [\"The email has been successfully sent to vinoglora.i+free@zohotest.com with distribution of cost views.\"] }}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open distribution of cost view\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"OPEN\", \"arguments\" : { \"module\" : \"VIEW\", \"name\" : \"distribution of cost\", \"type\" : \"view\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"1000030000005346634\", \"name\" : \"distribution of cost\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"export this view\",\n  \"output\" : [ \n      {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"distribution of cost\",\"type\":\"view\", \"viewid\":[\"1000030000005346634\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open distribution of cost report.\",\n    \"extractedOutput\" : [ \n        {  \"action\" : \"OPEN\", \"arguments\" : { \"module\" : \"REPORT\", \"name\" : \"distribution of cost\", \"type\" : \"report\" } } \n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"1000030000005346634\", \"name\" : \"distribution of cost\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"schedule this as png to abinesh.b@zohocorp.com every TUESDAY 5PM\",\n  \"output\" : [ \n       {\"action\":\"SCHEDULE\",\"arguments\":{\"module\":\"MAIL\",\"name\":\"Distribution of Cost\",\"type\":\"report\",\"viewid\":[\"1000030000005346634\"], \"exporttype\": [\"IMG\"], \"username\":[\"abinesh.b@zohocorp.com\"],\"time\":[\"17:00\"],\"scheduletype\":[\"WEEKLY\"],\"scheduleday\":[\"3\"],\"successmessage\":[\"The email has been successfully scheduled! It will be sent to abinesh.b@zohocorp.com every Tuesday at 5 PM.\"],\"schedulername\":[\"Distribution of Cost Schedule\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open sales table .\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"TABLE\",\"name\":\"sales\",\"type\":\"table\"}}      ]\n}, {\n    \"query\" : \"open distribution of cost report.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"distribution of cost\",\"type\":\"report\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"100003000000000221\",\"name\":\"Distribution of Cost\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"TABLE\":[{\"type\":\"Table\",\"viewid\":\"100003000000000009\",\"name\":\"sales table\"}]} \n  ]\n}, {\n  \"user_query\" : \"export as csv\",\n  \"output\" : [ \n       {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"Distribution of Cost\",\"type\":\"report\",\"viewid\":[\"100003000000000221\"],\"exporttype\":[\"CSV\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open sales table.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"TABLE\",\"name\":\"sales\",\"type\":\"table\"}}       ]\n}, {\n    \"query\" : \"region wise product category.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"region wise product category\",\"type\":\"report\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"NLPUNSAVEDCONFIG_21314738_1751600349970_0#100003000000000009\",\"name\":\"\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"TABLE\":[{\"type\":\"Table\",\"viewid\":\"100003000000000009\",\"name\":\"sales table\"}]} \n  ]\n}, {\n  \"user_query\" : \"export as pdf\",\n  \"output\" : [ \n       {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"region wise product category\",\"type\":\"report\",\"viewid\":[\"NLPUNSAVEDCONFIG_21314738_1751600349970_0#100003000000000009\"],\"exporttype\":[\"PDF\"]}}    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open cost metric analysis dashboard.\",\n    \"extractedOutput\" : [ \n           {\"action\":\"OPEN\",\"arguments\":{\"module\":\"DASHBOARD\",\"name\":\"cost metric analysis\",\"type\":\"dashboard\"}}       ]\n}, {\n    \"query\" : \"sales by region.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"sales by region\",\"type\":\"report\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"NLPUNSAVEDCONFIG_21314738_1751600005567_0#100003000000000009\",\"name\":\"\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"Dashboard\",\"viewid\":\"100003000000004317\",\"name\":\"Cost Metric Analysis Dashboard\"}]} \n  ]\n}, {\n  \"user_query\" : \"export as csv\",\n  \"output\" : [ \n       {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"sales by region\",\"type\":\"report\",\"viewid\":[\"NLPUNSAVEDCONFIG_21314738_1751600005567_0#100003000000000009\"],\"exporttype\":[\"CSV\"]}}    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"share distribution of cost report to @manoj.gattam@gmail.com as csv .\",\n    \"extractedOutput\" : [ \n           {\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"distribution of cost\",\"type\":\"report\",\"exporttype\":[\"CSV\"],\"username\":[\"manoj.gattam@gmail.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"], \"emailname\": [\"Distribution of Cost Update\"], \"successmessage\": [\"The email has been successfully sent to manoj.gattam@gmail.com with distribution of cost reports.\"] }}      ]\n}, {\n    \"query\" : \"open sales by region dashboard.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"DASHBOARD\",\"name\":\"sales by region\",\"type\":\"dashboard\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"Dashboard\",\"viewid\":\"100003000000000221\",\"name\":\"sales by region\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"100003000000000009\",\"name\":\"distribution of cost\"}]} \n  ]\n}, {\n  \"user_query\" : \"schedule this as pdf at 7PM\",\n  \"output\" : [ \n           {\"action\":\"SCHEDULE\",\"arguments\":{\"module\":\"MAIL\",\"name\":\"sales by region\",\"type\":\"dashboard\",\"viewid\":[\"100003000000000221\"],\"exporttype\":[\"PDF\"],\"time\":[\"19:00\"],\"schedulername\":[\"Sales for every region update Schedule\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"schedule distribution of cost to ravi@zohotest.com as csv at 12PM .\",\n    \"extractedOutput\" : [ \n{ \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"distribution of cost\", \"type\": \"view\", \"exporttype\": [\"CSV\"], \"username\": [\"ravi@zohotest.com\"], \"time\": [\"12:00\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to manoj.gattam@zohotest.com at 12 PM.\"] } }\n      ]\n}, {\n    \"query\" : \"open profit analysis report.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"profit analysis\",\"type\":\"report\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"100003000000000221\",\"name\":\"profit analysis\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"Dashboard\",\"viewid\":\"100003000000000009\",\"name\":\"distribution of cost\"}]} \n  ]\n}, {\n  \"user_query\" : \"share the above view\",\n  \"output\" : [ \n       {\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"profit analysis\",\"type\":\"report\",\"viewid\":[\"100003000000000221\"],\"sharetype\":[\"analytics\"]}} \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"share sales table to sandeep@gmail.com.\",\n    \"extractedOutput\" : [ \n           {\"action\":\"SHARE\",\"arguments\":{\"module\":\"TABLE\",\"name\":\"sales\",\"type\":\"table\",\"username\":[\"sandeep@gmail.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"], \"emailname\": [\"Sales table Update\"], \"successmessage\": [\"The email has been successfully sent to sandeep@gmail.com with sales table.\"]}}      ]\n}, {\n    \"query\" : \"open customer insights report.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"customer insights\",\"type\":\"report\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"100003000000000221\",\"name\":\"customer insights\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"TABLE\":[{\"type\":\"Table\",\"viewid\":\"100003000000000009\",\"name\":\"sales\"}]} \n  ]\n}, {\n  \"user_query\" : \"schedule this as csv at 8PM\",\n  \"output\" : [ \n           {\"action\":\"SCHEDULE\",\"arguments\":{\"module\":\"MAIL\",\"name\":\"customer insights\",\"type\":\"report\",\"exporttype\":[\"CSV\"],\"viewid\":[\"100003000000000221\"],\"time\":[\"20:00\"],\"schedulername\":[\"Customer Insights Schedule\"]}}    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"share customer insights report to peter.ramesh@gmail.com.\",\n    \"extractedOutput\" : [ \n           {\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"customer insights\",\"type\":\"report\",\"username\":[\"peter.ramesh@gmail.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"],\"emailname\": [\"Customer Insights Update\"], \"successmessage\": [\"The email has been successfully sent to peter.ramesh@gmail.com with customer insights reports.\"]}}\n      ]\n}, {\n    \"query\" : \"open cost analysis metric dashboard.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"DASHBOARD\",\"name\":\"cost analysis metric\",\"type\":\"dashboard\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"Dashboard\",\"viewid\":\"100003000000000221\",\"name\":\"cost analysis metric\"}]} \n  ]\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"100003000000000009\",\"name\":\"customer insights\"}]} \n  ],\n}, {\n  \"user_query\" : \"schedule this as csv at 10PM\",\n  \"output\" : [ \n           {\"action\":\"SCHEDULE\",\"arguments\":{\"module\":\"MAIL\",\"name\":\"cost analysis metric\",\"type\":\"dashboard\",\"exporttype\":[\"CSV\"],\"viewid\":[\"100003000000000221\"],\"time\":[\"22:00\"],\"schedulername\":[\"Cost Analysis Metric Schedule\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"schedule geo table to ramesh@test.com on every tuesday 7PM as csv.\",\n    \"extractedOutput\" : [ \n           { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"geo\", \"type\": \"table\", \"exporttype\": [\"CSV\"], \"username\": [\"ramesh@test.com\"], \"time\": [\"19:00\"], \"scheduletype\": [\"WEEKLY\"], \"scheduleday\": [\"3\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to ramesh@test.com every Tuesday at 7 PM.\"], \"schedulername\": [\"Sales Table Schedule\"] } } ]      ]\n}, {\n    \"query\" : \"open sales analysis dashboard.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"DASHBOARD\",\"name\":\"sales analysis\",\"type\":\"dashboard\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"dashboard\",\"viewid\":\"100003000000000221\",\"name\":\"sales analysis\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"TABLE\":[{\"type\":\"Table\",\"viewid\":\"100003000000000009\"}]} \n  ]\n}, {\n  \"user_query\" : \"share this as csv\",\n  \"output\" : [ \n       {\"action\":\"SHARE\",\"arguments\":{\"module\":\"DASHBOARD\",\"name\":\"sales analysis\",\"type\":\"dashboard\",\"viewid\":[\"100003000000000221\"],\"exporttype\":[\"CSV\"],\"sharetype\":[\"analytics\"]}}    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open new.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"new\",\"type\":\"view\"}}      ]\n}, {\n    \"query\" : \"open geo table.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"TABLE\",\"name\":\"geo\",\"type\":\"table\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"TABLE\":[{\"type\":\"table\",\"viewid\":\"100003000000000221\",\"name\":\"geo\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"Dashboard\",\"viewid\":\"100003000000000009\",\"name\":\"new\"}]} \n  ]\n}, {\n  \"user_query\" : \"export as csv\",\n  \"output\" : [ \n       {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"TABLE\",\"name\":\"geo\",\"type\":\"table\",\"viewid\":[\"100003000000000221\"],\"exporttype\":[\"CSV\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open cost metrics query table.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"QUERYTABLE\",\"name\":\"cost metrics\",\"type\":\"querytable\"}}      ]\n}, {\n    \"query\" : \"open distribution report.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"distribution\",\"type\":\"report\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"Report\",\"viewid\":\"100003000000000221\",\"name\":\"distribution\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"QUERYTABLE\":[{\"type\":\"querytable\",\"viewid\":\"100003000000000009\",\"name\":\"cost metrics\"}]} \n  ]\n}, {\n  \"user_query\" : \"export as pdf\",\n  \"output\" : [ \n       {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"distribution\",\"type\":\"report\",\"viewid\":[\"100003000000000221\"],\"exporttype\":[\"PDF\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open profit analysis dashboard .\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"DASHBOARD\",\"name\":\"profit analysis\",\"type\":\"dashboard\"}}      ]\n}, {\n    \"query\" : \"open distribution of cost view.\",\n    \"extractedOutput\" : [ \n       {\"action\":\"OPEN\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"distribution of cost\",\"type\":\"view\"}}      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       {\"REPORT\":[{\"type\":\"report\",\"viewid\":\"100003000000000221\",\"name\":\"Distribution of Cost\"}]} \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       {\"DASHBOARD\":[{\"type\":\"dashboard\",\"viewid\":\"100003000000000009\",\"name\":\"profit analysis\"}]} \n  ]\n}, {\n  \"user_query\" : \"export as csv\",\n  \"output\" : [ \n       {\"action\":\"EXPORT\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"Distribution of Cost\",\"type\":\"view\",\"viewid\":[\"100003000000000221\"],\"exporttype\":[\"CSV\"]}}\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export employee attendance summary views.\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"employee attendance\", \"type\": \"summaryview\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"REPORT\" : [ { \"type\" : \"summaryview | Report\", \"viewid\" : \"10000300000238752\"} ] } \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"export the first SUMMARYVIEW chart| EXPORT THIS | EXPORT IT .\",\n  \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"SUMMARYVIEW\", \"name\": \"employee attendance\", \"type\": \"summaryview\", \"viewid\": [\"10000300000238752\"] } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales analysis pivots\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales analysis\", \"type\": \"pivot\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"REPORT\" : [ { \"type\" : \"pivot | Report\", \"viewid\" : \"10000300000238752\"}, { \"type\" : \"pivot | Report\", \"viewid\" : \"10000300000223627\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"filter by year| filter by region | filter by cost.\",\n  \"output\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales analysis year | sales analysis region | sales analysis cost\", \"type\": \"pivot\" } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export profit dashboard\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"profit sales analysis\", \"type\": \"Dashboard\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"10000300000238752\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n       \n  ]\n}, {\n  \"user_query\": \" include insights dashboards also. | include revenue also\",\n  \"output\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"profit \", \"type\": \"dashboard\" } },\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \" insights | revenue\", \"type\": \"dashboard\" } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"search for view titled 'geo'\",\n    \"extractedOutput\" : [ \n           {\"action\":\"LIST\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"geo\",\"type\":\"view\"}}\n      ]\n},\n {\n  \"user_query\" : \"search for view titled 'geo' and 'sales vs cost\",\n  \"output\" : [ \n           {\"action\":\"LIST\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"geo\",\"type\":\"view\"}},{\"action\":\"LIST\",\"arguments\":{\"module\":\"VIEW\",\"name\":\"sales vs cost\",\"type\":\"view\"}}\n      ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export profit dashboard\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"profit sales analysis\", \"type\": \"Dashboard\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"10000300000238752\"} ] } \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n} ], {\n  \"user_query\": \"share sales vs cost report to dinesh.\",\n  \"output\" : [ \n        {\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"sales vs cost\",\"type\":\"report\",\"username\":[\"dinesh\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"], \"emailname\": [\"Sales vs Cost Update\"], \"successmessage\": [\"The email has been successfully sent to dinesh with sales vs cost reports.\"]}} \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales by region chart\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales by region\", \"type\": \"chart\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [{ \"type\" : \"report\", \"viewid\" : \"10000300000238752\"}, { \"type\" : \"report\", \"viewid\" : \"10000300000238752\"},{ \"type\" : \"AnalysisView|report\", \"viewid\" : \"10000300000238752\"}] } \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n ]\n}, {\n  \"user_query\": \" include customer insights also. | include product category also\",\n  \"output\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales by region \", \"type\": \"chart\" } },\n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"customer insights | product category\", \"type\": \"chart\" } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales by region chart as pdf\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales by region\", \"type\": \"chart\", \"exporttype\": [\"PDF\"] } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"report\", \"viewid\" : \"100003000002387214\" },{ \"type\" : \"report\", \"viewid\" : \"10000300000238752\" },{ \"type\" : \"AnalysisView|report\", \"viewid\" : \"100003000002323548\" } ] } \n  ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\": \" export the above charts as csv. | export them as csv | export the three charts as CSV\",\n  \"output\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales by region\", \"type\": \"chart\", \"exporttype\": [\"CSV\"], \"viewid\" : [\"100003000002387214\", \"10000300000238752\", \"100003000002323548\"] } } \n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export customer feedback chart as pdf\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"customer feedback\", \"type\": \"chart\", \"exporttype\": [\"PDF\"] } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"report\", \"viewid\" : \"100003000002387214\" },{ \"type\" : \"AnalysisView|report\", \"viewid\" : \"100003000002323548\" } ] \n    } \n  ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\": \" export first and third charts as csv. \",\n  \"output\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"customer feedback\", \"type\": \"chart\", \"exporttype\": [\"CSV\"], \"viewid\" : [\"100003000002387214\", \"100003000002323548\"] } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export profit dashboard as pdf\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"profit\", \"type\": \"dashboard\", \"exporttype\": [\"PDF\"] } }\n      ]\n  }, {\n    \"query\" : \"filter by region wise. \",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"profit region wise\", \"type\": \"dashboard\", \"exporttype\": [\"PDF\"] } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"100003000002387214\"}, { \"type\" : \"Dashboard\", \"viewid\" : \"10000300000238752\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\": \"export the first and second as pdf file | export the above as pdf file\",\n  \"output\" : [ \n      {  \"action\" : \"EXPORT\", \"arguments\" : { \"module\" : \"DASHBOARD\", \"name\" : \"profit region wise\", \"type\" : \"dashboard\", \"viewid\":[\"100003000002387214\",\"10000300000238752\"], \"exporttype\": [\"PDF\"] } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export customer dashboard\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"customer\", \"type\": \"dashboard\" } }\n      ]\n  }, {\n    \"query\" : \"export the first and second as csv\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"customer\", \"viewid\": [\"100004000000094004\",\"100003000002387214\" ], \"type\": \"dashboard\",  \"exporttype\": [\"CSV\"]} }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n       { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"100004000000094004\"}, { \"type\" : \"Dashboard\", \"viewid\" : \"100003000002387214\"} ] } \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\": \"create a mail schedule using the first dashboard as html to @manoj.gattam on every wednesday at evening 7 PM\",\n  \"output\" : [ \n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"customer\", \"viewid\": [\"100004000000094004\"], \"type\": \"dashboard\" , \"exporttype\": [\"HTML\"], \"username\": [\"manoj.gattam\"], \"time\": [\"19:00\"], \"scheduletype\": [\"WEEKLY\"], \"scheduleday\": [\"4\"], \"scheduleinterval\": [\"1\"],  \"successmessage\": [\"The email has been successfully scheduled! It will be sent to manoj.gattam on every Wednesday at 8 AM.\"], \"schedulername\": [\"Customer Insights Analysis Schedule\"] }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export customer dashboard\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"customer\", \"type\": \"dashboard\" } }\n      ]\n  },\n  {\n    \"query\" : \"export sales pivots\",\n    \"extractedOutput\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales\", \"type\": \"pivot\" } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Pivot | Report\", \"viewid\" : \"100004000000092555\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n    { \"Dashboard\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"100004000000094004\"}, { \"type\" : \"Dashboard\", \"viewid\" : \"100003000002387214\"} ] }\n  ]\n}, {\n  \"user_query\": \"share this to @manoj.gattam\",\n  \"output\" : [ \n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"PIVOT\", \"name\": \"sales\", \"viewid\": [\"100004000000092555\"], \"type\": \"pivot\",\"username\": [\"manoj.gattam\"] , \"usertype\": [\"person\"], \"sharetype\": [\"analytics\"],\"emailname\": [\"Sales Analysis Update\"], \"successmessage\": [\"The email has been successfully sent to manoj.gattam with sales reports.\"]} }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"sales by region\",\n    \"extractedOutput\" : \"sales by region\"\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ {\"type\":\"Report\",\"viewid\":\"NLPUNSAVEDCONFIG_53546928_1749711619584_0#44064000021543029\",\"name\":\"\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\": \"export this chart\",\n  \"output\" : [ \n    { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales by region\", \"viewid\": [\"NLPUNSAVEDCONFIG_53546928_1749711619584_0#44064000021543029\"], \"type\": \"chart\" } } \n  ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open sales dashboard\",\n    \"extractedOutput\" : [ \n        { \"action\": \"OPEN\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"sales\", \"type\": \"dashboard\" } }\n      ]\n  },   {\n    \"query\" : \"open customer insights report\",\n    \"extractedOutput\" : [ \n        { \"action\": \"OPEN\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"customer insights\", \"type\": \"report\" } }\n      ]\n  } ]\n},{\n  \"currentFinalConsolidatedOutput\" : [ \n    { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"1000040000000942435\"} ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n    { \"DASHBOARD\" : [ { \"type\" : \"Dashboard\", \"viewid\" : \"100004000000092555\"} ] }\n  ]\n} , {\n  \"user_query\": \"export the above dashboard as excel.\",\n  \"output\" : [ \n        { \"action\": \"EXPORT\", \"arguments\": { \"module\": \"DASHBOARD\", \"name\": \"sales\", \"viewid\": [\"100004000000092555\"], \"type\": \"dashboard\", \"exporttype\": [\"XLS\"]} }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open sales chart\",\n    \"extractedOutput\" : [ \n        { \"action\": \"OPEN\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales\", \"type\": \"chart\" } }\n      ]\n  },\n  {\n    \"query\" : \"schedule the above chart as csv to manoj | schedule it as csv to manoj.\",\n    \"extractedOutput\" : [ \n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"sales\", \"type\": \"chart\", \"viewid\":[\"100004000000092555\"], \"exporttype\": [\"CSV\"], \"username\":[\"manoj\"], \"usertype\":[\"person\"], \"schedulername\":[\"Sales Performance Analysis\"], \"successmessage\":[\"The email has been successfully scheduled! It will be sent to manoj.\"] } }\n      ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n      { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"100004000000092555\" } ] }\n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\": \"7 PM\",\n  \"output\" : [ \n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"sales\", \"type\": \"report\", \"viewid\":[\"100004000000092555\"], \"exporttype\": [\"CSV\"], \"username\":[\"manoj\"], \"usertype\":[\"person\"], \"time\": [\"19:00\"], \"schedulername\":[\"Sales Performance Analysis Schedule\"], \"successmessage\":[\"The email has been successfully scheduled! It will be sent to manoj at 7 PM.\"] } }\n    ]\n} ],\n[{\"previousQueryDetails\":[{\"query\":\"open sales chart\",\"extractedOutput\":[{\"action\":\"OPEN\",\"arguments\":{\"module\":\"CHART\",\"name\":\"sales\",\"type\":\"chart\"}}]},{\"query\":\"export the above chart as csv\",\"extractedOutput\":[{\"action\":\"EXPORT\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"Cost vs Sales\",\"type\":\"report\",\"viewid\":[\"100003000000002258\"],\"exporttype\":[\"CSV\"]}}]},{\"query\":\"share it to abinesh.b@zohocorp.com\",\"extractedOutput\":[{\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"Cost vs Sales\",\"type\":\"report\",\"viewid\":[\"100003000000002258\"],\"username\":[\"abinesh.b@zohocorp.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"],\"emailname\":[\"Cost vs Sales Update\"],\"successmessage\":[\"The email has been successfully sent to abinesh.b@zohocorp.com with Cost vs Sales reports.\"]}}]}]},{\"currentFinalConsolidatedOutput\":[{\"REPORT\":[{\"type\":\"report\",\"viewid\":\"100003000000002258\",\"name\":\"Cost vs Sales\"}]}]},{\"previousFinalConsolidatedOutput\":[]}, {\n  \"user_query\": \"send this as mail\",\n  \"output\" : [ \n        { \"action\": \"SHARE\", \"arguments\": { \"module\": \"REPORT\", \"name\": \"Cost vs Sales\", \"type\": \"report\", \"viewid\":[\"100004000000092555\"], \"exporttype\": [\"CSV\"],\"sharetype\":[\"mail\"], \"username\":[\"abinesh.b@zohocorp.com\"], \"usertype\":[\"person\"], \"emailname\":[\"Cost vs Sales Update\", \"successmessage\":[\"The email has been successfully sent to abinesh.b@zohocorp.com with Cost vs Sales reports.\"] } }\n    ]\n} ],\n[{\"previousQueryDetails\":[{\"query\":\"open sales chart\",\"extractedOutput\":[{\"action\":\"OPEN\",\"arguments\":{\"module\":\"CHART\",\"name\":\"sales\",\"type\":\"chart\"}}]},{\"query\":\"export the above chart as csv\",\"extractedOutput\":[{\"action\":\"EXPORT\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"Cost vs Sales\",\"type\":\"report\",\"viewid\":[\"100003000000002258\"],\"exporttype\":[\"CSV\"]}}]},{\"query\":\"share it to abinesh.b@zohocorp.com\",\"extractedOutput\":[{\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"Cost vs Sales\",\"type\":\"report\",\"viewid\":[\"100003000000002258\"],\"username\":[\"abinesh.b@zohocorp.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"],\"emailname\":[\"Cost vs Sales Update\"],\"successmessage\":[\"The email has been successfully sent to abinesh.b@zohocorp.com with Cost vs Sales reports.\"]}}]}]},{\"currentFinalConsolidatedOutput\":[{\"REPORT\":[{\"type\":\"report\",\"viewid\":\"100003000000002258\",\"name\":\"Cost vs Sales\"}]}]},{\"previousFinalConsolidatedOutput\":[]}, {\n  \"user_query\": \"schedule this at 10AM\",\n  \"output\" : [ \n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"Cost vs Sales\", \"type\": \"report\", \"viewid\":[\"100004000000092555\"], \"exporttype\": [\"CSV\"], \"username\":[\"abinesh.b@zohocorp.com\"], \"usertype\":[\"person\"], \"time\": [\"10:00\"],\"schedulername\":[\"Cost Vs Sales Analysis Schedule\"], \"successmessage\":[\"The email has been successfully scheduled with cost vs sales reports! It will be sent to abinesh.b@zohocorp.com at 10 PM.\"] } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open sales chart\",\n    \"extractedOutput\" : [ \n        { \"action\": \"OPEN\", \"arguments\": { \"module\": \"CHART\", \"name\": \"sales\", \"type\": \"chart\" } }\n      ]\n  } ]\n},\n  \"finalConsolidatedOutput\" : [ \n  { \"REPORT\" : [ { \"type\" : \"Report\", \"viewid\" : \"100004000000092555\"}, { \"type\" : \"Report\", \"viewid\" : \"1000040000000926556\"}, { \"type\" : \"Report\", \"viewid\" : \"10000400000067657\"} ]\n  } ]\n}, {\n  \"user_query\": \"schedule it as csv to manoj on Monday at 8PM\",\n  \"output\" : [ \n        { \"action\": \"SCHEDULE\", \"arguments\": { \"module\": \"MAIL\", \"name\": \"sales\", \"type\": \"chart\", \"time\": [\"20:00\"], \"viewid\": [\"100004000000092555\", \"1000040000000926556\", \"10000400000067657\"], \"exporttype\": [\"CSV\"], \"username\": [\"manoj\"], \"usertype\": [\"person\"], \"schedulername\": [\"Sales Performance Analysis Schedule\"], \"successmessage\": [\"The email has been successfully scheduled! It will be sent to manoj on Monday at 8 PM.\"] } }\n    ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales view\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"VIEW\",\"name\" : \"sales\",\"type\" : \"view\"}\n    } ]\n  } ]\n}, {\n  \"previousFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172326\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172146\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172154\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172126\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172118\"}]\n  } ]\n}, {\n  \"user_query\" : \"export pivot chart\",\n  \"output\" : [ \n    {\"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"type\" : \"report\"} }\n   ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales report\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : { \"module\" : \"REPORT\",\"name\" : \"sales\",\"type\" : \"report\"}\n    } ]\n  }, {\n    \"query\" : \"export pivot chart\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"type\" : \"pivot\"}\n    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ \n  ],\n  \"previousFinalConsolidatedOutput\" : [ \n  ]\n}, {\n  \"user_query\" : \"cost\",\n  \"output\" :[ \n    {\"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"name\" : \"cost\",\"type\" : \"pivot\"} }\n   ]\n} ][ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export sales report\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : { \"module\" : \"REPORT\",\"name\" : \"sales\",\"type\" : \"report\"}\n    } ]\n  }, {\n    \"query\" : \"export pivot chart\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"type\" : \"pivot\"}\n    } ]\n  }, {\n    \"query\" : \"cost\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"name\" : \"cost\",\"type\" : \"pivot\"}\n    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"Pivot | Report\",\"viewid\" : \"1000030000002452423\"} ]\n  } ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\" : \"share it to sandeep.kotra\",\n  \"output\" :[ {\n    \"action\" : \"SHARE\",\"arguments\" : {\"module\" : \"PIVOT\",\"name\" : \"cost\",\"type\" : \"pivot\", \"viewid\" : [\"1000030000002452423\"], \"username\": [\"sandeep.kotra\"], \"usertype\": [\"person\"],\"sharetype\": [\"analytics\"], \"emailname\": [\"Cost Update\"], \"successmessage\": [\"The email has been successfully sent to sandeep.kotra with cost pivot views.\"] }\n  } ]\n} ][ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"export pivot charts\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"type\" : \"pivot\"}\n      } ]\n  }, {\n    \"query\" : \"sales\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"PIVOT\",\"name\" : \"sales\",\"type\" : \"pivot\"}\n    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"Pivot | Report\",\"viewid\" : \"1000030000002452423\"} ]\n  } ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\" : \"send it as email to @manoj.gattam include insights also\",\n  \"output\" : [ \n    {\"action\" : \"SHARE\",\"arguments\" : {\"module\" : \"PIVOT\",\"name\" : \"sales\",\"type\" : \"pivot\",\"viewid\" : [\"1000030000002452423\"],\"username\" : [ \"manoj.gattam\" ],\"usertype\" : [ \"person\" ],\"sharetype\" : [ \"mail\" ], \"emailname\": [\"Sales Update\"], \"successmessage\": [\"The email has been successfully sent to manoj.gattam with sales reports.\"]}\n   ]\n} ][ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"list all table\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"LIST\",\"arguments\" : {\"module\" : \"TABLE\",\"type\" : \"table\"}\n    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"TABLE\" : [ {\"type\" : \"Table\",\"viewid\" : \"100003000000172326\"} ]\n  } ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\" : \"export the above table.\",\n  \"output\" : [ {\n    \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"TABLE\",\"name\" : \"name\",\"type\" : \"table\", \"viewid\" : [\"100003000000172326\"]}\n  } ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"schedule distribution of cost as excel to manoj.gattam@zohotest.com on august, sep and oct 5th\",\n    \"extractedOutput\" : [ {\n       {\"action\":\"SCHEDULE\",\"arguments\":{\"module\":\"MAIL\",\"name\":\"distribution of cost\",\"type\":\"view\", \"exporttype\": [\"XLS\"],\"username\":[\"manoj.gattam@zohotest.com\"],\"scheduletype\":[\"YEARLY\"],\"scheduleday\":[\"5\"],\"scheduleinterval\":[\"1\"],\"schedulemonth\":[\"7\",\"8\",\"9\"],\"successmessage\":[\"The email has been successfully scheduled! It will be sent to manoj.gattam@zohotest.com on 5th August, September, and October.\"],\"schedulername\":[\"Distribution of Cost Update Schedule\"]}}     } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"Pivot | Report\",\"viewid\" : \"1000030000002452423\"} ]\n  } \n  \"previousFinalConsolidatedOutput\" : [ ]\n}], {\n  \"user_query\" : \"10am\",\n  \"output\" : [ {\n       {\"action\":\"SCHEDULE\",\"arguments\":{\"module\":\"MAIL\",\"name\":\"distribution of cost\",\"type\":\"view\", \"viewid\" : [\"1000030000002452423\"],\"exporttype\": [\"XLS\"],\"username\":[\"manoj.gattam@zohotest.com\"],\"time\":[\"10:00\"],\"scheduletype\":[\"YEARLY\"],\"scheduleday\":[\"5\"],\"scheduleinterval\":[\"1\"],\"schedulemonth\":[\"7\",\"8\",\"9\"],\"successmessage\":[\"The email has been successfully scheduled! It will be sent to manoj.gattam@zohotest.com at 10 AM on 5th August, September, and October.\"],\"schedulername\":[\"Distribution of Cost Update Schedule\"]}}  } ]\n} ],\n[ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"list all reports\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"LIST\",\"arguments\" : {\"module\" : \"REPORT\",\"type\" : \"report\"}\n    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172326\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172154\"} ]\n  } ],\n  \"previousFinalConsolidatedOutput\" : [ {\n  } ]\n} ], {\n  \"user_query\" : \"export the first and third report.\",\n  \"output\" : [ \n    {\"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"REPORT\",\"name\":\"name\", \"type\" : \"report\", \"viewid\" : [\"100003000000172326\", \"100003000000172154\"]}\n  ]\n} ] [ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"share sales report to manoj.gattam@gmail.com\",\n    \"extractedOutput\" : [ {\n       {\"action\":\"SHARE\",\"arguments\":{\"module\":\"REPORT\",\"name\":\"sales\",\"type\":\"report\",\"username\":[\"manoj.gattam@gmail.com\"],\"usertype\":[\"person\"],\"sharetype\":[\"analytics\"], \"emailname\": [\"sales Update\"], \"successmessage\": [\"The email has been successfully sent to manoj.gattam@gmail.com with sales reports.\"]}}    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000172326\"}]\n  } ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\" : \"schedule this as pdf everyday\",\n  \"output\" : [ {\n    \"action\" : \"SCHEDULE\",\n    \"arguments\" : {\"module\" : \"MAIL\",\"name\" : \"sales\",\"type\" : \"report\",\"viewid\" : [\"100003000000172326\"], \"exporttype\": [\"PDF\"],\"username\" : [ \"manoj.gattam@gmail.com\" ],\"scheduletype\" : [ \"DAILY\" ],\"successmessage\" : [ \"The email has been successfully scheduled! It will be sent to manoj.gattam@zohocorp.com every day.\" ],\"schedulername\":[\"Sales Performance Analysis Schedule\"]}   } ]\n} ][ {\n  \"previousQueryDetails\" : [ {\n    \"query\" : \"open sales report\",\n    \"extractedOutput\" : [ {\n      \"action\" : \"OPEN\",\"arguments\" : {\"module\" : \"REPORT\",\"name\" : \"sales\",\"type\" : \"report\"}\n    } ]\n  } ]\n}, {\n  \"currentFinalConsolidatedOutput\" : [ {\n    \"REPORT\" : [ {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000296003\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"1000030000002645344\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"10000300000236743\"} ]\n  } ],\n  \"previousFinalConsolidatedOutput\" : [ ]\n}, {\n  \"user_query\" : \"also send this as mail\",\n  \"output\" : [ {\n    \"action\" : \"SHARE\",\"arguments\" : {\"module\" : \"REPORT\",\"name\" : \"sales\",\"type\" : \"report\",\"viewid\" : [ \"100003000000296003\" ],\"sharetype\" : [ \"mail\" ]}\n  } ]\n} ],[{\n    \"previousQueryDetails\" : [ {\n      \"query\" : \"export cost and sales view\",\n      \"extractedOutput\" : [ \n      { \"action\" : \"EXPORT\",\"arguments\" : { \"module\" : \"VIEW\",\"name\" : \"cost\",\"type\" : \"view\"}},\n      { \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"VIEW\",\"name\" : \"sales\",\"type\" : \"view\"}}\n    ]},\n     {\n      \"query\" : \"export first three reports\",\n      \"extractedOutput\" : [ {\n        \"action\" : \"EXPORT\",\"arguments\" : {\"module\" : \"REPORT\",\"viewid\" : [ \"100003000000253211\", \"10000300005345325\", \"10000300000124124\" ], \"type\" : \"report\"}\n      } ]\n    } ]\n  }, \n  {\n    \"currentFinalConsolidatedOutput\" : [\n       {\"REPORT\" : [ {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"100003000000253211\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"10000300005345325\"}, {\"type\" : \"AnalysisView | Report\",\"viewid\" : \"10000300000124124\"} ]}\n    ],\n    \"previousFinalConsolidatedOutput\" : [\n       {\"DASHBOARD\" : [ {\"type\" : \"Dashboard\",\"viewid\" : \"1000030000003124534\"}, {\"type\" : \"Dashboard\",\"viewid\" : \"100003000001241435\"}, {\"type\" : \"Dashboard\",\"viewid\" : \"10000300001244523\"} ]}\n    ]\n  }, \n  {\n    \"user_query\" : \"also share them to dinesh.ravindran@gmail.com\",\n    \"output\":[ \n        {\"action\" : \"SHARE\",\"arguments\" : {\"module\" : \"REPORT\",\"name\":\"cost\", \"viewid\" : [ \"100003000000253211\", \"10000300005345325\", \"10000300000124124\" ],\"type\" : \"report\",\"username\" : [ \"dinesh.ravindran@gmail.com\" ],\"usertype\" : [ \"person\" ],\"sharetype\" : [ \"analytics\" ], \"emailname\": [\" Cost and Sales Update\"], \"successmessage\": [\"The email has been successfully sent to dinesh.ravindran@gmail.com with cost and sales reports.\"]}}\n      ]\n  }] ]}"}

za.askzia.chart_narration.qwen.prompt.for.zia.turbo=You are an assistant that generates natural, user-friendly narrations of analytical queries.\n\nInputs:\n1. Query Interpretation → structured description of the query (dimensions, measures, criteria, functions, ranking).\n2. Query Initial narration -> the initial narration generated for the query.\n\nSynonyms:\nDimensions = Categories, Attributes, Columns\nMeasures = Metrics, Values, Numbers\nRanking/Criteria = Sorting, Conditions, Filters\nNarration = Summary of the query Interpretation\n\nGuidelines:\n- Use the Query Interpretation as the **primary source of truth** to build the narration.\n- The output must not be a closely similar version of the initial narration.\n. Use Initial narration only as a reference to understand user intent.\n- Clearly describe the dimension(s), measure(s), and ranking/criteria in simple business language.\n- Keep narration short and user-friendly sentence.\n- Never miss any columns or filters from the Query Interpretation.\n- Prefer number digits over words for quantities (e.g., "5" instead of "five").\n- `Forecast` is a keyword use it only if its is forecast query (`forecast` implies projection into future based on historical data).\n- Use the word `Forecast` only when `forecast` is used in the Query Initial narration.\n- Use the word `Trend` only when `trend` is used in the Query Initial narration.\n\nExpected Tone:\n- Forecast : Total <measure> forecasted for <date range> based on <date range>\n\nCorrect Example:\n- Date in years.\n- Customers List.\n- All Provinces.\n- The Region with the lowest Total Sales.\n- The Customer Segment with the highest Average Discount.\n- Total Revenue by month over the past year.\n- Total revenue by Product Category.\n- Average delivery time by Ship Mode.\n- Monthly loss across all Regions.\n- Percentage of returned orders by Product.\n- Yearly revenue by region.\n- Total Sales contribution by top 5 Products.\n- Total Profit forecasted for the next 6 months based on past 2 years data.\n\nWrong Examples:\n- Total Sales by month with forecasted sales for the next 12 months based on the last 3 years of data.\n- Sales by Region where the province is United States of America.\n- Total Sales by Region for each month and year, showing the trend with 100% confidence.\n\nHow to capitalize:\n- Always capitalize the first letter of the narration.\n- Always capitalize proper nouns (e.g., "United States", "January").\n- Always capitalize the first letter of Column names in Query Interpretation.\n\nRight Examples:\nTotal Sales by Region.\nPercentage of returned orders by Product.\nTotal Matches won by India in the year 2023.\n\nWrong Examples:\ntotal sales by region. (first letter not capitalized, column names (region, sales) not capitalized)\npercentage of returned orders by product. (First letter and column name (product) not capitalized)\ntotal matches won by india in the year 2023. (First letter, proper noun (India) and column name (matches) not capitalized)\n\nWords Strictly NOT to be used:\n- Dimension, Measure, Function, Bottom 1, Top 1, Bottom N, Top N, SQL-like clauses, Filter, Filtered, Group by, Order by, Show me, Display, Chart, Graph, Visualization, Trend, Plot, Draw, Drawn, Plotted, confidence.\n\nOutput format:\n{\n  "narration": "<natural-language summary here>",\n}

za.askzia.chart_narration_simple.qwen.prompt.for.zia.turbo=You are an assistant that generates natural, user-friendly narrations of analytical queries.\n\nInputs:\n1. Query Interpretation → structured description of the query (dimensions, measures, criteria, functions, ranking).\n2. Query Initial narration -> the initial narration generated for the query.\n3. User Query → the original user query.\n\nSynonyms:\nMeasures = Metrics, Values, Numbers\nNarration = Summary of the query Interpretation\n\nGeneral Guidelines Guidelines:\n- Use the Query Initial narration as the **primary source of truth** to build the narration.\n- The output must not be a closely similar version of the initial narration.\n. Use Initial narration only as a reference to understand user intent.\n- Keep narration short and user-friendly sentence.\n- Never miss any columns or filters from the Query Interpretation.\n- Prefer number digits over words for quantities (e.g., "5" instead of "five").\n- Pick nuances from user query only if they align with Query Interpretation.\n- Use column synonyms when they offer clearer or more meaningful context for the chart results.\n\nHow to Handle Data:\n- Query Initial narration is the primary source of truth.\n- It will contain columns, filters, functions and data.\n- Simplify the filters and function names if required.\n- Use column synonyms found `Query Interpretation` to improve clarity.\n\nEg. Initial Query Narration: "Number of Faculty Expertise with Faculty Expertise as Civil Engineering is 7958"\nGood Output:  "Number of Faculty Expertise in Civil Engineering is 7958."\nBad Output: "Number of Faculty Expertise in Civil Engineering" (Missing the data value)\n\nEg. Initial Query Narration: "Total Sales where the order-data is January 2023 is 1500000"\nGood Output: "Total Sales in January 2023 is 1,500,000 across"\nBad Output: Total Sales the order-data is January 2023 is 1500000" (missed using better alias)\n\nEg. Initial Query Narration: "Average Salary where faculties department is Computer Science and Year of Joining is 2020 is 75290.59"\nGood Output: "Average Salary in Computer Science department for the year 2020 is 75,290.59"\nBad Output: "Average Salary in Computer Science department for the year 2020 is 11000" (Changed the value)\n\nHow to capitalize:\n- Always capitalize the first letter of the narration.\n- Always capitalize proper nouns (e.g., "United States", "January").\n- Always capitalize the first letter of Column names in Query Interpretation.\n\nRight Examples:\nTotal Sales and Profit is 102.282.\nPercentage of returned orders by Product.\nTotal Matches won by India in the year 2023.\n\nWrong Examples:\ntotal sales by region. (first letter not capitalized, column names (region, sales) not capitalized)\npercentage of returned orders by product. (First letter and column name (product) not capitalized)\ntotal matches won by india in the year 2023. (First letter, proper noun (India) and column name (matches) not capitalized)\n\nWords Strictly NOT to be used:\n- Dimension, Measure, Function, Bottom 1, Top 1, Bottom N, Top N, SQL-like clauses, Filter, Filtered, Group by, Order by, Show me, Display, Chart, Graph, Visualization, Trend, Plot, Draw, Drawn, Plotted, confidence.\n\nOutput format:\n{\n  "narration": "<natural-language summary here>",\n}

# Agent Prompts
za.askzia.report_creation_agent.narration.prompt=You are an assistant that generates natural, user-friendly narrations of analytical queries. Your goal is to convert structured query interpretations into short, clear, and business-friendly crisp one-liner that sound natural to end users. Each narration should read like a simple observation or statement about the data — concise, accurate, and easy to understand without any technical jargon.\n\nGoal:\nConvert the structured Query Interpretation into a single, clear, and business-friendly sentence that reads like a natural observation of the data. The narration must be concise, human-sounding, and never exceed one line.\n\nInputs:\n1. Query Interpretation → structured description of the query (dimensions, measures, criteria, functions, ranking).\n\nSynonyms:\nDimensions = Categories, Attributes, Columns\nMeasures = Metrics, Values, Numbers\nRanking/Criteria = Sorting, Conditions, Filters\n\nMandatory Rules:\n- Use the Query Interpretation as the primary source of truth to build the narration.\n- Output must be exactly one sentence and fit in a single line.\n- Narration should sound like a simple business statement, not an explanation or report.\n- Clearly describe the dimension(s), measure(s), and ranking/criteria in simple business language. \n- Keep narration short and user-friendly sentence.\n- Never miss any columns or filters from the Query Interpretation.\n- Prefer number digits over words for quantities (e.g., \"5\" instead of \"five\").\n- Use the word \"Forecast\" only when the query involves forecast (projection into the future based on historical data).\n- Use the word \"Trend\" only when the query involves trend analysis.\n- Provide a concise, descriptive title of the result (1 sentence or phrase).\n- Base this strictly on the resolved columns, tables, and measures present in the final JSON.\n- Do not infer or include columns/measures not in the resolved output.\n- Always capitalize:\n    - The first letter of the narration.\n    - Proper nouns (e.g., \"United States\", \"January\")\n    - Column names from the Query Interpretation.\n\nExpected Output Format:\n{\n  \"narration\": \"<natural-language one line narration>\"\n}\n\nTone:\n- Friendly, concise, and business-oriented.\n- Avoid any technical or SQL-like phrasing.\n- Use only the columns, tables, and measures resolved for the current query.\n- Do not propose other analyses, columns, groupings, or future actions.\n- Do not speculate about trends or features not present in the current result.\n- Clearly mention the key metric and dimension from the actual result.\n- Frame it as a natural observation (e.g., \"You can see…\" or \"This shows…\" or any other contextually relevant variations) rather than a suggestion.\n- Base this only on the query interpretation data, not on the user query alone.\n\nStrict Rules:\n- Never use words: \"Dimension\", \"Measure\", \"Function\", \"Bottom 1\", \"Top 1\", \"Bottom N\", \"Top N\", \"SQL\", \"Filter\", \"Filtered\", \"Group by\", \"Order by\", \"Show me\", \"Display\", \"Chart\", \"Graph\", \"Visualization\", \"Trend\", \"Plot\", \"Draw\", \"Drawn\", \"Plotted\", \"confidence\".\n- Do not mention statistical or visualization context.\n- Narration must sound like a natural summary of what a business user would ask or see.\n- Do not add full stop at the end\n- Strictly do not return long descriptions/explanations\n\n\n\nCorrect Examples:\n- Total Sales by Region.\n- Percentage of returned orders by Product.\n- Total Matches won by India in the year 2023.\n- Total Profit forecasted for the next 6 months based on past 2 years data.\n- Average Delivery Time by Ship Mode.\n- Total Revenue by month over the past year.\n- Total Sales contribution by top 5 Products.\n\nIncorrect Examples:\n- total sales by region. (missing capitalization)\n- percentage of returned orders by product. (missing capitalization)\n- total matches won by india in the year 2023. (proper noun not capitalized)\n- Total Sales by Region for each month and year, showing the trend with 100% confidence. (unwanted words)\n- Sales by Region where the province is United States of America. (wordy and unnatural)
za.askzia.json_validator.prompt.for.zia.turbo=ROLE & PURPOSE:\nyou are a strict JSON formatter. Your ONLY task is to validate and convert the given input into the exact JSON structure specified below.\n\nCRITICAL RULES:\n 1. Output MUST be valid JSON\n2. Output MUST follow the exact structure shown below\n3. Do NOT add explanations, markdown formatting, or any text outside the JSON\n4. Do NOT use code blocks or backticks\n5. If any field is missing in the input, infer it logically or set appropriate defaults\n6. Preserve all data from the input while conforming to the structure.\n\nREQUIRED JSON STRUCTURE:\n{   \"needClarity\": true or false,   \"promptUserInput\": \"HTML block as string\",   \"narration\": \"description as string\",   \"visualizationType\": \"chart type as string\",   \"dimensions\": [array of dimension objects],   \"measures\": [array of measure objects],   \"criteria\": [array of criteria objects],   \"concept\": [array],   \"follow_up_questions\": [\"Q1\", \"Q2\", \"Q3\"] }\n\nDIMENSION OBJECT FORMAT:\n{   \"sortType\": \"ASCENDING\" or \"DESCENDING\" or \"NONE\",   \"isTable\": true or false,   \"columnName\": \"string\",   \"tableName\": \"string\",   \"function\": \"actual\" or \"date enum\" or \"other\",   \"axisPosition\": \"x-axis\" or \"y-axis\" or \"group\" or \"tooltip\" or \"color-axis\" or \"text-axis\" }\n\nMEASURE OBJECT FORMAT: {   \"sortType\": \"ASCENDING\" or \"DESCENDING\" or \"NONE\",   \"isTable\": false,   \"columnName\": \"string\",   \"tableName\": \"string\",   \"function\": \"sum\" or \"count\" or \"avg\" or other aggregation function,   \"windowFunction\": \"running total\" or other window function or empty string,   \"windowFunctionDimensionIndex\": 0 or valid index,   \"axisPosition\": \"x-axis\" or \"y-axis\" or other valid position }\n\nCRITERIA OBJECT FORMAT: {   \"isCriteriaIncluded\": true or false,   \"columnName\": \"string\",   \"tableName\": \"string\",   \"function\": \"month\" or \"year\" or \"sum\" or \"actual\" or other function,   \"criteriaType\": \"TEXT\" or \"DATE\" or \"NUMERIC\" or \"RANKING\",   \"criteriaValues\": [\"value1\", \"value2\"],   \"isFutureDate\": true or false }\n \nOUTPUT ONLY THE VALID JSON. NO OTHER TEXT.
za.askzia.json_validator.qwen.prompt.for.zia.turbo=ROLE & PURPOSE:\nyou are a strict JSON formatter. Your ONLY task is to validate and convert the given input into the exact JSON structure specified below.\n\nCRITICAL RULES:\n 1. Output MUST be valid JSON\n2. Output MUST follow the exact structure shown below\n3. Do NOT add explanations, markdown formatting, or any text outside the JSON\n4. Do NOT use code blocks or backticks\n5. If any field is missing in the input, infer it logically or set appropriate defaults\n6. Preserve all data from the input while conforming to the structure.\n\nREQUIRED JSON STRUCTURE:\n{   \"needClarity\": true or false,   \"promptUserInput\": \"HTML block as string\",   \"narration\": \"description as string\",   \"visualizationType\": \"chart type as string\",   \"dimensions\": [array of dimension objects],   \"measures\": [array of measure objects],   \"criteria\": [array of criteria objects],   \"concept\": [array],   \"follow_up_questions\": [\"Q1\", \"Q2\", \"Q3\"] }\n\nDIMENSION OBJECT FORMAT:\n{   \"sortType\": \"ASCENDING\" or \"DESCENDING\" or \"NONE\",   \"isTable\": true or false,   \"columnName\": \"string\",   \"tableName\": \"string\",   \"function\": \"actual\" or \"date enum\" or \"other\",   \"axisPosition\": \"x-axis\" or \"y-axis\" or \"group\" or \"tooltip\" or \"color-axis\" or \"text-axis\" }\n\nMEASURE OBJECT FORMAT: {   \"sortType\": \"ASCENDING\" or \"DESCENDING\" or \"NONE\",   \"isTable\": false,   \"columnName\": \"string\",   \"tableName\": \"string\",   \"function\": \"sum\" or \"count\" or \"avg\" or other aggregation function,   \"windowFunction\": \"running total\" or other window function or empty string,   \"windowFunctionDimensionIndex\": 0 or valid index,   \"axisPosition\": \"x-axis\" or \"y-axis\" or other valid position }\n\nCRITERIA OBJECT FORMAT: {   \"isCriteriaIncluded\": true or false,   \"columnName\": \"string\",   \"tableName\": \"string\",   \"function\": \"month\" or \"year\" or \"sum\" or \"actual\" or other function,   \"criteriaType\": \"TEXT\" or \"DATE\" or \"NUMERIC\" or \"RANKING\",   \"criteriaValues\": [\"value1\", \"value2\"],   \"isFutureDate\": true or false }\n \nOUTPUT ONLY THE VALID JSON. NO OTHER TEXT.
