Feedoptimise

Returns detailed analysis data for a specific feed's last processing run and historical trends. Includes: total item count, number of blocked/excluded items, analyser results with breakdowns (e.g. missing prices, invalid URLs, title quality), and processing statistics over time. Use this to understand feed health, identify data quality issues, or check how many items passed through analysers. Requires a feed_id — use feeds_list to find the feed_id if unknown.

get_feed_analysis

Retrieves a specific product as it appears in a particular feed's output, after all feed-level mapping, modifiers, and transformations have been applied. If the product has variants, variant-level field values override parent-level values (variant 'wins'). Use this to see exactly what data is being sent to a channel (Google Shopping, Meta, etc.) for a given product. Requires both feed_id and item_id. Use feeds_list to find the feed_id and item_search to find the item_id if not already known. To see the raw source data before feed transformations, use get_product instead.

get_feed_product

Retrieves full details of a product as it exists in the source (imported raw data). Returns all mapped fields, modifier outputs, analyser results, override values, variant data, and metadata (date added, date updated). Use this to inspect the original imported data before any feed-specific transformations. Requires a product/item ID. To see how the same product looks after feed-level mapping and transformations, use get_feed_product instead.

get_product

Fetches and parses metadata from a product's landing page URL. The landing page URL is resolved from the product data using source_id and product_id. Returns structured data such as schema.org metadata, Open Graph tags, Meta/Facebook pixel data, Google tag data, and other embedded metadata. Use this to verify whether landing page data matches feed data, such as price, availability, or condition. Requires source_id and product_id. Use sources_list to find source_id and item_search to find product_id if unknown.

get_product_page_metadata

Returns detailed configuration for a specific import source. Includes: connector type and settings (Shopify, Magento, WooCommerce, URL, FTP, SFTP, API), full mapping schema (field mappings, modifiers, variables), import schedule (times, frequency, days), backup/drop-protection settings, status (active/paused), and analyser configurations. Use this to inspect how data is imported and transformed at source level. For the full mapping schema of a single source, do NOT request schema in sources_list — instead call this tool with the source_id obtained from sources_list.

get_source

Returns detailed analysis data for a specific source's last import run and historical trends. Includes: total imported item count, number of blocked/excluded items, analyser results with breakdowns (e.g. missing fields, invalid data, brand distribution), and processing statistics over time. Use this to understand source data health, identify import issues, or check how many items were processed and excluded. Requires a source_id — use sources_list to find it if unknown.

get_source_analysis

Returns the complete operational instructions, platform context, and tool usage guidelines for the Feedoptimise MCP server. IMPORTANT: Call this tool FIRST at the start of every new conversation before using any other tools. The response contains essential context about the Feedoptimise platform architecture (sources, feeds, catalog, mapping, modifiers, analysers), the data flow model, tool chaining patterns, safety rules, and best practices. Without this context, tool calls may be incorrect or miss important dependencies. This tool is for agent behavior, not for detailed platform tutorials. For any how-to question, manual, setup guide, modifier explanation, mapping tutorial, onboarding flow, or feature usage explanation, use `articles_search` as the primary source of truth because it performs semantic search across the platform knowledge base and tutorials.

instruction

Retrieves Google Merchant Center diagnostic details for a single product as connected through a specific feed. Returns item-level issues such as disapproval reasons, attribute warnings, and policy violations from GMC. Requires feed_id and product_id. Use feeds_list to find feed_id and item_search to find product_id if unknown.

item_gmc_diagnosis

Searches the product catalog for items matching the query. Supports searching by: title, brand, item ID, MPN, SKU, GTIN, or variant data. Returns a list of matching products with basic info (ID, title, brand, price, status, image). Use this as the starting point when a user asks about a specific product — the returned item_id can then be used with get_product, get_feed_product, block_source_product, block_feed_product, or item_gmc_diagnosis.

item_search

Returns a list of all Google Merchant Center accounts connected to the current Feedoptimise account. Each entry includes the GMC account ID and name. Use this before calling gmc_diagnosis or item_gmc_diagnosis to identify the correct GMC account. If multiple GMC accounts are returned, ask the user to specify which one (in the internal agent, ui_confirm_selection would be used; in MCP context, return the list for the client to handle selection).

list_gmcs

Triggers an immediate full import (force update) of product data from the connected store/platform into the source. This pulls fresh data from the external system (Shopify, Magento, WooCommerce, URL feed, etc.), replacing the current source data. After completion, all feeds connected to this source will automatically sync. Use this when: the user has updated their store data and wants it reflected immediately, or when troubleshooting stale data issues. Note: this is a 'Force Update' (full re-download from store). If only Feedoptimise mapping/rules changed and store data hasn't changed, a reprocess may be more appropriate and faster. Requires a source_id — use sources_list to find it if unknown.

source_force_import

Returns a summary list of all import sources in the account. Each entry includes: source_id, source name, campaign_id, status (active/paused/error), connector type (Shopify, Magento, WooCommerce, URL, FTP, SFTP, etc.), whether it is active, and last run timestamp. Use this to discover available sources, find a specific source_id for use with other tools (get_source, source_force_import, get_source_analysis, etc.), or get a quick overview of all source statuses. For detailed configuration of a specific source, call get_source with the source_id. When only identifying sources, set include_schema to false. Set include_schema to true only when the user specifically asks for source mapping or schema details.

sources_list

List all Feedoptimise accounts the authenticated user has access to.

Set the active account for this MCP session. Required before using other tools if the user has multiple accounts.