get_available_filters
ChatGPTRetrieves all available filter columns the current user has access to, contextualized by location and product. Required Tool Calling Order: 1. get_locations_and_products_overview: Discover high-level category names. 2. lookup_location_and_product_ids: Get the exact numeric location_ids and product_ids. 3. `get_available_filters` (THIS TOOL): Pass the discovered IDs here to see which dynamic filters (e.g., Brand, Retailer) apply to this specific product/location combination. 4. get_filter_options: Discover valid exact string values for the columns returned here. 5. get_filtered_data: Finally, fetch the actual volume/time-series data.
get_available_filters
ChatGPTRetrieves all available filter columns the current user has access to, contextualized by location and product. Required Tool Calling Order: 1. get_locations_and_products_overview: Discover high-level category names. 2. lookup_location_and_product_ids: Get the exact numeric location_ids and product_ids. 3. `get_available_filters` (THIS TOOL): Pass the discovered IDs here to see which dynamic filters (e.g., Brand, Retailer) apply to this specific product/location combination. 4. get_filter_options: Discover valid exact string values for the columns returned here. 5. get_filtered_data: Finally, fetch the actual volume/time-series data.
get_brands_and_retailers
ChatGPTRetrieves available Brands and Retailers, optionally filtered by Location or Product IDs. LLM Usage Instructions: 1. Overview vs. Search: If you do not provide a brand_search or retailer_search, you will only receive a truncated view (max 250) of the available data. To find specific items, you MUST provide these search terms. 2. Step 2 (Optional): Call this tool AFTER lookup_location_and_product_ids if the user specifies an interest in particular brands or retailers. 3. Filtering (Optional): Pass location_ids or product_ids obtained from Step 1 to narrow down the list of relevant brands/retailers for a specific market segment. 4. Data Extraction: Use the returned string values to populate the brands or retailers arguments in the get_data tool.
get_brands_and_retailers
ChatGPT[DEPRECATED] This tool is strictly for legacy system compatibility. New agents MUST use the get_available_filters tool instead. Retrieves available Brands and Retailers, optionally filtered by Location or Product IDs. LLM Usage Instructions: 1. Overview vs. Search: If you do not provide a brand_search or retailer_search, you will only receive a truncated view (max 250) of the available data. To find specific items, you MUST provide these search terms. 2. Step 2 (Optional): Call this tool AFTER lookup_location_and_product_ids if the user specifies an interest in particular brands or retailers. 3. Filtering (Optional): Pass location_ids or product_ids obtained from Step 1 to narrow down the list of relevant brands/retailers for a specific market segment. 4. Data Extraction: Use the returned string values to populate the brands or retailers arguments in the get_data tool.
get_data
ChatGPTFetches the actual time-series volume data for market analysis. LLM Usage Instructions: 1. Optional Filters: All filters (location, product, brand, retailer) and datasets are OPTIONAL. 2. Dataset Selection: If the user wants to explore specific data views, provide the appropriate dataset(s) in the datasets parameter: - Use ["Strategische Dataset - Merk"] when the user wants to explore data of brands. - Use ["Strategische Dataset - Retailer"] when the user wants to explore data of retailers. - Use ["Strategische Dataset - Total"] when the user wants to explore products, product details or get an overall market view. 3. Interval: Specify whether to aggregate the data on a "monthly" or "yearly" basis using the interval parameter. 4. Metrics: Control the output metrics by toggling include_volumes, include_growth_mom, and include_growth_yoy. 5. Prerequisites (Only if filtering): IF the user asks to filter by specific locations, products, brands, or retailers, use get_locations_and_products_overview and/or get_brands_and_retailers first to retrieve the correct IDs/names. 6. Execution: If the user wants an overall market view without specific filters, call this tool directly without passing any arguments.
get_data
ChatGPT[DEPRECATED] This tool is strictly for legacy system compatibility. New agents MUST use the get_filtered_data tool instead. Fetches the actual time-series volume data for market analysis. LLM Usage Instructions: 1. Optional Filters: All filters (location, product, brand, retailer) and datasets are OPTIONAL. 2. Dataset Selection: If the user wants to explore specific data views, provide the appropriate dataset(s) in the datasets parameter: - Use ["Strategische Dataset - Merk"] when the user wants to explore data of brands. - Use ["Strategische Dataset - Retailer"] when the user wants to explore data of retailers. - Use ["Strategische Dataset - Total"] when the user wants to explore products, product details or get an overall market view. 3. Interval: Specify whether to aggregate the data on a "monthly" or "yearly" basis using the interval parameter. 4. Metrics: Control the output metrics by toggling include_volumes, include_growth_mom, and include_growth_yoy. 5. Prerequisites (Only if filtering): IF the user asks to filter by specific locations, products, brands, or retailers, use get_locations_and_products_overview and/or get_brands_and_retailers first to retrieve the correct IDs/names. 6. Execution: If the user wants an overall market view without specific filters, call this tool directly without passing any arguments.
get_filter_options
ChatGPTRetrieves the available, specific filter values for up to 5 filter columns, scoped to the given location and product. Required Tool Calling Order: 1. get_locations_and_products_overview: Discover high-level category names. 2. lookup_location_and_product_ids: Get the exact numeric product_ids. 3. get_available_filters: Discover valid filter columns for these IDs. 4. `get_filter_options` (THIS TOOL): Pass up filter column names, PLUS the same product_ids, to see what values you can filter by. 5. get_filtered_data: Finally, fetch the actual volume/time-series data using these options.
get_filter_options
ChatGPTRetrieves the available, specific filter values for up to 5 filter columns, scoped to the given location and product. Required Tool Calling Order: 1. get_locations_and_products_overview: Discover high-level category names. 2. lookup_location_and_product_ids: Get the exact numeric product_ids. 3. get_available_filters: Discover valid filter columns for these IDs. 4. `get_filter_options` (THIS TOOL): Pass up filter column names, PLUS the same product_ids, to see what values you can filter by. 5. get_filtered_data: Finally, fetch the actual volume/time-series data using these options.
get_filtered_data
ChatGPTFetches time-series volume data with support for dynamic filtering across multiple strategic datasets. Dataset Descriptions (`datasets` argument): - "Strategische Dataset - Total": Baseline dataset for high-level, unsegmented overall trends. - "Strategische Dataset - Merk": Use for brand-specific data (e.g., Nike, Samsung). - "Strategische Dataset - Retailer": Use for store/vendor-specific data (e.g., Bol.com, Amazon). - "Strategische dataset - Specifications": Use for product features, sizes, colors, or technical specs. - "Strategische dataset - Customer Segment": Use for audience demographics or user segments (e.g., children, women). - "Strategische dataset - Purpose": Identifies the specific use-case, environment, or gifting occasion driving the purchase. - "Strategische dataset - Information": Use for analyzing what information needs consumers have. - "Strategische dataset - Purchase Intention": Use for commercial intent signals and buying readiness. LLM Usage Instructions: 1. Overviews & Comparisons (CHOOSE A DATASET): In order to get an overview or compare different dimensions (such as comparing different brands, retailers, or specifications against each other), you MUST explicitly choose the corresponding dataset from the list above and pass it via the datasets argument. 2. Optional Specific Filtering (USE APPLIED FILTERS): In order to filter the dataset on specific dimensions (e.g., isolating the data to only show "Nike" or only "Sneakers"), you MUST use the applied_filters argument. - Use get_available_filters and get_filter_options to discover appropriate filters. - Example applied_filters payload: { "brand": ["nike", "adidas"], "category": ["sneakers"] } 3. Optional Location & Product Context: Use get_locations_and_products_overview and lookup_location_and_product_ids to find relevant location_ids and product_ids. 4. Defaults: All filters (location, product, applied_filters) and datasets are OPTIONAL. If no dataset is explicitly provided, the tool defaults to "Total".
get_filtered_data
ChatGPTFetches time-series volume data with support for dynamic filtering across multiple strategic datasets. Dataset Descriptions (`datasets` argument): - "Strategische Dataset - Total": Baseline dataset for high-level, unsegmented overall trends. - "Strategische Dataset - Merk": Use for brand-specific data (e.g., Nike, Samsung). - "Strategische Dataset - Retailer": Use for store/vendor-specific data (e.g., Bol.com, Amazon). - "Strategische dataset - Specifications": Use for product features, sizes, colors, or technical specs. - "Strategische dataset - Customer Segment": Use for audience demographics or user segments (e.g., children, women). - "Strategische dataset - Purpose": Identifies the specific use-case, environment, or gifting occasion driving the purchase. - "Strategische dataset - Information": Use for analyzing what information needs consumers have. - "Strategische dataset - Purchase Intention": Use for commercial intent signals and buying readiness. LLM Usage Instructions: 1. Overviews & Comparisons (CHOOSE A DATASET): In order to get an overview or compare different dimensions (such as comparing different brands, retailers, or specifications against each other), you MUST explicitly choose the corresponding dataset from the list above and pass it via the datasets argument. 2. Optional Specific Filtering (USE APPLIED FILTERS): In order to filter the dataset on specific dimensions (e.g., isolating the data to only show "Nike" or only "Sneakers"), you MUST use the applied_filters argument. - Use get_available_filters and get_filter_options to discover appropriate filters. - Example applied_filters payload: { "brand": ["nike", "adidas"], "category": ["sneakers"] } 3. Optional Location & Product Context: Use get_locations_and_products_overview and lookup_location_and_product_ids to find relevant location_ids and product_ids. 4. Defaults: All filters (location, product, applied_filters) and datasets are OPTIONAL. If no dataset is explicitly provided, the tool defaults to "Total".
get_locations_and_products_overview
ChatGPTRetrieves a top-level overview of the master hierarchy for Locations and Products (Categories). 1. Call this tool FIRST to understand the available categories without cluttering the context with IDs. 2. Once you find the relevant names from this overview, use the lookup_location_and_product_ids tool to fetch the actual IDs and subcategories.
get_locations_and_products_overview
ChatGPTRetrieves a top-level overview of the master hierarchy for Locations and Products (Categories). 1. Call this tool FIRST to understand the available categories without cluttering the context with IDs. 2. Once you find the relevant names from this overview, use the lookup_location_and_product_ids tool to fetch the actual IDs and subcategories.
lookup_location_and_product_ids
ChatGPTRetrieves specific IDs from categories using lookup terms. LLM Usage Instructions: 1. Call this AFTER using get_locations_and_products_overview. 2. Provide a location_lookup or product_lookup based on the overview to fetch location_ids and product_ids. 3. Traverse subOptions in the response to find the specific subcategories or states ids.
lookup_location_and_product_ids
ChatGPTRetrieves specific IDs from categories using lookup terms. LLM Usage Instructions: 1. Call this AFTER using get_locations_and_products_overview. 2. Provide a location_lookup or product_lookup based on the overview to fetch location_ids and product_ids. 3. Traverse subOptions in the response to find the specific subcategories or states ids.
