mcpField
N
n8n API MCP Server

n8n API MCP Server

Local
github.com/jasondsmith72Developer Tools1Updated 83w ago

About n8n API MCP Server

An MCP server for interacting with self-hosted n8n instances through the n8n API. It combines endpoint search from a local OpenAPI index with reusable calls stored in Fast Memory for natural-language lookup.

Key features

  1. 01Execute n8n API calls
  2. 02Search OpenAPI endpoints
  3. 03Inspect endpoint parameters and responses
  4. 04Find API calls with natural language
  5. 05Cache and manage successful calls

Use cases

  1. 01Find workflow API operations in natural language
  2. 02Reuse previously validated API calls

MCP support

Exposes tools for searching, inspecting, and executing n8n API calls to MCP clients. It connects to an n8n instance configured with its URL and API key.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

No install method listed yet

We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.

Repository

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

n8n API MCP Server

A Model Context Protocol (MCP) server designed to interact with a self-hosted n8n instance.

This server provides tools to:

  • Execute calls to the n8n API.
  • Search for n8n API endpoints using a local database populated from an OpenAPI specification.
  • Store and retrieve frequently used or successful API calls using a "Fast Memory" database for quicker natural language access.

Features

Databases
  • API Specification Database (api_spec.db): Stores details about n8n API endpoints, loaded from an OpenAPI/Swagger JSON file. Used by search and detail tools.
  • Fast Memory Database (fast_memory.db): Caches successful API calls linked to natural language queries. Allows for quick retrieval of previously used calls and suggests saving new successful calls.
Tools
  • search_api_endpoints: Search the local API spec database for endpoints matching a query (path, summary, description, tags).
  • get_api_endpoint_details: Retrieve detailed information (parameters, request body, responses) for a specific endpoint (path and method) from the local API spec database.
  • execute_api_call: Execute an API call to the configured n8n instance. Checks Fast Memory first; if not found, prompts to save the call to Fast Memory upon success.
  • natural_language_api_search: Search for API calls using natural language. Checks Fast Memory first, then falls back to searching the API spec database based on the query.
  • save_to_fast_memory: Manually save an API call (path, method, params, data) along with its original natural language query and a description to the Fast Memory database.
  • list_fast_memory: List entries currently stored in the Fast Memory database, optionally filtering by a search term.
  • delete_from_fast_memory: Delete a specific entry from the Fast Memory database by its ID.
  • clear_fast_memory: Remove all entries from the Fast Memory database.
  • load_api_spec_from_json: Load or update the API Specification database (api_spec.db) by parsing an OpenAPI/Swagger JSON file.
  • send_raw_api_request: Execute an API call using a raw request string (e.g., "GET /workflows?limit=5"). Uses the same underlying logic as execute_api_call, including Fast Memory checks and save prompts.

Setup

  1. Clone the repository:
    git clone https://github.com/jasondsmith72/N8N-api-MCP.git
    cd N8N-api-MCP
    
  2. Install dependencies:
    npm install
    
  3. Build the server:
    npm run build
    
    This compiles the TypeScript code to JavaScript in the build directory.

Configuration

This server requires environment variables to connect to your n8n instance. These are typically set in your MCP client's configuration file (e.g., cline_mcp_settings.json).

  • N8N_URL: The base URL of your n8n instance (e.g., http://your-n8n.example.com). Defaults to http://localhost:5678 if not set.
  • N8N_API_KEY: Your n8n API key for authentication.

Loading the API Specification

Before using the search tools (search_api_endpoints, natural_language_api_search), you need to populate the API Specification database.

  1. Obtain the OpenAPI (Swagger) JSON specification file for your n8n version. This might be available from the n8n documentation or potentially downloadable from your instance (e.g., /api/v1/docs-json).
  2. Use the load_api_spec_from_json tool, providing the absolute path to the downloaded JSON file:
    load_api_spec_from_json(json_file_path="C:\\path\\to\\your\\n8n-openapi.json")
    

Installation for Cline / Claude Desktop

Add the following configuration to your cline_mcp_settings.json (adjust the path to index.js based on your clone location):

{
  "mcpServers": {
    "... other servers ...": {},
    "n8n-api-mcp": {
      "command": "node",
      "args": [
        "C:\\MCP-Servers\\N8N-api-MCP\\build\\index.js" // <-- Update this path if needed
      ],
      "env": {
        "N8N_URL": "YOUR_N8N_INSTANCE_URL", // e.g., "http://192.168.1.100:5678"
        "N8N_API_KEY": "YOUR_N8N_API_KEY_HERE"
      },
      "disabled": false,
      "autoApprove": []
    }
  }
}

(Remember to replace YOUR_N8N_INSTANCE_URL and YOUR_N8N_API_KEY_HERE with your actual values)

Development

For development with auto-rebuild on file changes:

npm run watch
Debugging

Use the MCP Inspector for easier debugging:

npm run inspector

The Inspector provides a web interface to view MCP communication.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.