広告掲載
MCP.so
ログイン

n8n API MCP Server

@jasondsmith72

リポジトリ

基本情報

カテゴリ

開発者ツール

トランスポート

stdio

作者

jasondsmith72

n8n API MCP Server について

MCP Server for interacting with an n8n instance

#n8n#api#mcp#server

n8n API MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "n8n-api-mcp": {
      "command": "node",
      "args": [
        "C:\\MCP-Servers\\N8N-api-MCP\\build\\index.js"
      ],
      "env": {
        "N8N_URL": "",
        "N8N_API_KEY": ""
      }
    }
  }
}

n8n API MCP Server のツール

10

Search the local API spec database for endpoints matching a query (path, summary, description, tags).

Retrieve detailed information (parameters, request body, responses) for a specific endpoint (path and method) from the local API spec database.

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.

Search for API calls using natural language. Checks Fast Memory first, then falls back to searching the API spec database based on the query.

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 entries currently stored in the Fast Memory database, optionally filtering by a search term.

Delete a specific entry from the Fast Memory database by its ID.

Remove all entries from the Fast Memory database.

Load or update the API Specification database (`api_spec.db`) by parsing an OpenAPI/Swagger JSON file.

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.

概要

What is n8n API MCP Server?

An MCP server that interacts with a self-hosted n8n instance, enabling API calls, searching endpoints via a local database populated from an OpenAPI spec, and storing frequently used calls in a Fast Memory database for quicker natural language access.

How to use n8n API MCP Server?

Clone the repository, install dependencies (npm install), build the server (npm run build), and configure environment variables (N8N_URL, N8N_API_KEY) in your MCP client’s settings (e.g., cline_mcp_settings.json). Then load the API spec using load_api_spec_from_json before using search tools.

Key features of n8n API MCP Server

  • Two local SQLite databases: API spec cache and Fast Memory.
  • Search endpoints by path, summary, description, or tags.
  • Execute API calls with automatic Fast Memory prompts.
  • Natural language search that checks memory first, then spec.
  • Manual save, list, delete, and clear Fast Memory entries.
  • Raw request string execution (send_raw_api_request).

Use cases of n8n API MCP Server

  • Automate workflow management via natural language queries to n8n.
  • Quickly reuse previously successful API calls without remembering details.
  • Explore and test n8n API endpoints without external documentation.
  • Build AI assistants that can interact with an n8n instance programmatically.

よくある質問

1

What dependencies or runtime are required?

Node.js, npm, and your n8n instance URL plus an API key are required. The server is built with TypeScript and uses SQLite local databases.

2

Where does the API specification data come from?

It is loaded from a downloaded OpenAPI/Swagger JSON file using the `load_api_spec_from_json` tool. This file may be available from your n8n instance at `/api/v1/docs-json`.

3

How does the Fast Memory database work?

It caches successful API calls linked to natural language queries. On a successful call via `execute_api_call` or `send_raw_api_request` (where the call is not already in Fast Memory), the server prompts to save it. It supports manual save, list, delete, and clear operations.

4

What transport does the server use?

The server communicates via standard MCP protocol but the README does not specify a particular transport layer (stdio or HTTP). Configuration examples show a `command: "node"` setup typical for stdio transport.

5

Are there any known limits?

The README does not mention any specific limits. The server is designed for self-hosted n8n instances and requires the user to provide the URL and API key.

コメント

「開発者ツール」の他のコンテンツ

Yocoolab

Yocoolab

MCP server for AI Agents — exposes Yocoolab feedback threads and design selections as tools

1

Capgo MCP

Cap-go

Console, Backend and CLI to manage Capgo Instant update and Native build for Capacitor apps Build and upload to app store your mobile app

Zyte MCP

Zyte

Zyte MCP is a remote MCP server available at https://mcp.zyte.com/v1/mcp/, that lets an agent use Zyte on your behalf: to get web pages, to extract data from them, to search the web, and to work with your Scrapy Cloud pr

consult

neelsatyavolu

Get a second opinion on your code from every coding agent you have. Claude Code, Codex and Grok review each other's diffs and answer each other's questions, even inside a session that's already running. Uses the CLIs you

NextBlock CMS

NextBlock

Open-source Next.js + Supabase CMS with a per-site MCP server: scaffold a site, then create pages, posts, navigation, media and SEO from Claude, Cursor or any MCP client. Streamable HTTP with a bearer token you mint in y

広告掲載