eBay Catalog API Tool Server for AI Agents (MCP) - n8n Workflow

• Developers and engineers building AI agents (e.g., using LangChain) who need reliable, structured access to eBay catalog data.


• E-commerce analysts requiring automated data feeds from eBay for market research.


• n8n Automation Specialists seeking advanced n8n templates for tool calling architectures.


• Users looking to implement specialized webhook functionality using a dedicated n8n trigger.

The modern requirement for AI agents often involves giving them access to external systems via specialized tools. This advanced n8n workflow solves the problem of integrating complex APIs, like the eBay Catalog API, directly into an AI agent's operational toolkit. By utilizing the MCP (Multimodal Context Planner) n8n trigger, this entire n8n workflow transforms into a single, cohesive server endpoint.

This setup allows the AI to send natural language requests (like 'Find product details for EPID 12345') which the n8n node automatically translates into the correct, authenticated eBay API calls, extracting structured data. This capability is critical for building sophisticated AI-powered listing assistants or market data trackers. This specific n8n template provides two core functions: retrieving product details by ePID and searching product summaries using various filters.

This automation operates as a constantly running server, triggered solely by external AI agent requests directed at the MCP endpoint.

1. AI Agent Request (n8n Trigger): The process starts when an external AI agent sends a request to the dedicated Catalog MCP Server n8n trigger node. This request includes the desired operation (tool name) and the necessary parameters, which are provided contextually by the AI.


2. Parameter Extraction ($fromAI()): The subsequent HTTP Request Tool n8n node nodes utilize the proprietary $fromAI() expression within their parameter fields. This expression dynamically pulls the required input (e.g., epid, q, X-EBAY-C-MARKETPLACE-ID) directly from the AI agent's request payload, ensuring that the API call is accurately structured based on the AI's intent.


3. API Execution: Depending on the tool called by the AI agent, either the Retrieve Product Details or Search Product Summaries n8n node executes an authenticated HTTP GET request to the eBay Catalog API.


4. Response: The HTTP Request Tool n8n node captures the raw JSON response from eBay and returns it directly back to the Catalog MCP Server n8n trigger, which then transmits the final result back to the initiating AI agent.

To deploy this powerful n8n workflow and make eBay data accessible to your AI agents, follow these steps:

1. Import the n8n Template: Copy the provided JSON and import it directly into your n8n instance using the 'New' -> 'Import from JSON' function.


2. Configure Credentials: The workflow requires OAuth2 credentials for the eBay API. You must set up a Generic Credential of type 'HTTP Header Auth' (or similar configuration compatible with eBay's OAuth 2.0 requirements) and ensure it is selected within the 'Authentication' section of both HttpRequestTool n8n node nodes.


3. Activate the n8n Workflow: Toggle the workflow status to 'Active'. Since the trigger is an MCP Server, activating it immediately makes the external webhook URL available.


4. Retrieve MCP URL: Copy the webhook URL displayed on the Catalog MCP Server n8n trigger node.


5. Connect to AI Agent: Use the copied MCP URL to configure the custom tool definition within your external AI agent framework (e.g., LangChain, custom orchestration layer). This effectively connects the AI agent to your n8n workflow.

Catalog MCP Server (n8n trigger):
Function: Acts as the entry point and server for AI agent tool-calling requests. It listens for HTTP POST requests defined by the MCP protocol.
Key Configuration: Configured with a specific path (/catalog-mcp) to serve as the webhook endpoint for the n8n automation.
Retrieve Product Details (HTTP Request Tool n8n node):
Function: Retrieves specific product details from the eBay Catalog API using the product's unique ePID.
Key Configuration: The URL path parameter epid is dynamically generated using {{ $fromAI('epid', ...) }}. Requires X-EBAY-C-MARKETPLACE-ID in the header, also derived from AI context.
Search Product Summaries (HTTP Request Tool n8n node):
Function: Searches for products based on keywords (q), category IDs, GTINs, or MPNs, returning summarized results.
* Key Configuration: All crucial query parameters (including q, aspectfilter, categoryids, and limit) are populated using $fromAI() expressions, allowing the AI agent to fully customize the search parameters dynamically through this specialized n8n node.