What does the MCP server do?
It enables Claude AI and Cursor IDE to interact with n8n via the Model Context Protocol, allowing you to create, manage, and monitor workflows using natural language commands.
n8n Workflow Builder MCP Server
@salacoste
基本情報
n8n Workflow Builder MCP Server について
AI-powered n8n workflow automation through natural language. MCP server enabling Claude AI & Cursor IDE to create, manage, and monitor workflows via Model Context Protocol. Multi-instance support, 17 tools, comprehensive docs. Build workflows conversationally without manual JSON
n8n Workflow Builder 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": {
"mcp-n8n-workflow-builder": {
"command": "npx",
"args": [
"@kernel.salacoste/n8n-workflow-builder",
"--version"
]
}
}
}n8n Workflow Builder MCP Server のツール
24List all workflows from n8n with essential metadata only (ID, name, status, dates, node count, tags). Supports filtering by active status. Optimized for performance to prevent large data transfers.
Manually execute a workflow by ID
Create a new workflow in n8n
Get a workflow by ID
Update an existing workflow
Partially update workflow (returns guidance - PATCH not supported by n8n API). Use update_workflow for actual updates.
Delete a workflow by ID
Activate a workflow by ID
Deactivate a workflow by ID
List all executions from n8n with optional filters
Get details of a specific execution by ID
Delete an execution by ID
Retry a failed workflow execution (creates new execution). Only works for executions with status "error".
Create a new tag
Get all tags
Get a tag by ID
Update a tag
Delete a tag
List all credentials (metadata only, sensitive data excluded for security). Returns credential ID, name, type, nodes access information, and timestamps. Actual credential data (passwords, tokens, keys) is never returned.
Get credential by ID (returns guidance - GET not supported by n8n API for security). Credentials contain sensitive data and cannot be read through REST API.
Update credential (returns guidance - PUT not supported by n8n API for security). Use delete + create pattern instead.
Create a new credential for external service authentication. Supports multiple credential types (httpBasicAuth, httpHeaderAuth, OAuth2, etc.). Use get_credential_schema to understand required fields for each type. Sensitive data is automatically encrypted by n8n.
Delete a credential by ID. Removes the credential from n8n permanently. Workflows using this credential will need to be updated.
Get JSON schema for a specific credential type (e.g., httpBasicAuth, googleDriveOAuth2Api). Returns field definitions, types, required fields, and validation rules to help create credentials programmatically.
概要
What is n8n Workflow Builder MCP Server?
An AI-powered MCP server that lets you build, manage, and monitor n8n workflows through natural language using Claude AI and Cursor IDE. It transforms workflow automation by enabling conversational creation and management of complex n8n workflows, eliminating manual JSON editing and UI navigation.
How to use n8n Workflow Builder MCP Server?
Install globally via npm (npm install -g @kernel.salacoste/n8n-workflow-builder). Configure with a .config.json (multi‑instance) or a .env file (single instance). Add the server to your Claude Desktop configuration (claude_desktop_config.json) or Cursor IDE configuration (.cursor/mcp.json). Restart the client and describe workflows in plain English.
Key features of n8n Workflow Builder MCP Server
- AI-powered workflow creation from natural language descriptions
- Multi-instance management (dev, staging, production) from one server
- 17 comprehensive tools covering workflows, executions, tags, and credentials
- Natural language interface – no JSON editing required
- Secure by design with API key encryption and credential protection
- 38+ documentation pages with guides, tutorials, and API reference
Use cases of n8n Workflow Builder MCP Server
- Build and test workflows locally, then promote through staging to production
- Monitor execution status and debug failed workflows across environments
- Export workflows from one n8n instance and import to another with automatic adaptation
- Generate workflow documentation and learn n8n patterns through AI guidance
よくある質問
What are the prerequisites?
You need Node.js v14+ (v18+ recommended), npm v7+, a running n8n instance with API access (tested with n8n v1.82.3+), and either Claude Desktop or Cursor IDE.
How does multi-instance management work?
A single MCP server can manage multiple n8n environments (e.g., production, staging, development) by reading a `.config.json` file that defines each environment’s host and API key. The server routes requests to the correct instance based on context.
What tools are included in the MCP server?
Eight workflow tools (create, update, delete, activate, execute, etc.), four execution tools (monitor, retry, delete), five tag management tools, and six credential tools (schema retrieval, creation, deletion – note that list/get/update credential are blocked by the n8n API and only provide guidance).
Is the MCP server secure?
Yes. It features built-in credential protection, API key encryption, secure multi-instance configuration, and never exposes sensitive data in logs.
「推論」の他のコンテンツ
Sandbox Mcp
pottekkatA Model Context Protocol (MCP) server that enables LLMs to run ANY code safely in isolated Docker containers.
118
iFlytek Workflow MCP Server
iflytekThis a simple implementation of an MCP server using iFlytek. It enables calling iFlytek workflows through MCP tools.
28
NeoCoder: Neo4j-Guided AI Coding Workflow
angrysky56An MCP server allowing AI assistants to use a Neo4j knowledge graph as their primary, dynamic instruction manual and long term project memory with adaptive templating and autonomous tool development tools.
20
MCP Advanced Reasoning Server for Cursor AI
AzDeltaQQThis is a very basic implementation of an Mcp-Reasoning-Server for Cursor AI .
13
Agentic Tools MCP Server
PimzinoA comprehensive Model Context Protocol (MCP) server providing AI assistants with powerful task management and agent memories capabilities with project-specific storage.
87
コメント