n8n MCP Server
Enables AI assistants to securely manage n8n workflows, executions, and credentials with enterprise-grade security features like TLS enforcement and audit logging.
README
n8n MCP Server
An enterprise-grade Model Context Protocol (MCP) server that enables AI assistants like Claude to securely interact with n8n workflow automation platform.
Built for regulated industries including cryptocurrency, fintech, and enterprises requiring strict security controls.
Security Features
- TLS/SSL Enforcement: HTTPS required by default, certificate verification enabled
- Audit Logging: Comprehensive logging with correlation IDs for compliance
- Rate Limiting: Built-in protection against abuse
- Input Validation: Strict validation with injection pattern detection
- Sensitive Data Redaction: API keys and secrets never logged
- Request Tracing: Correlation IDs for debugging and forensics
See SECURITY.md for full security documentation.
Features
- Workflow Management: Create, read, update, delete, activate, and deactivate workflows
- Execution Control: Execute workflows manually and monitor execution history
- Credential Visibility: List available credentials (metadata only)
- Health Monitoring: Check connectivity and server status
Prerequisites
- Python 3.10 or higher
- An n8n instance with API access enabled
- n8n API key (How to generate)
Quick Start
1. Clone the repository
git clone https://github.com/tommytang2414/n8n-mcp-server.git
cd n8n-mcp-server
2. Install dependencies
pip install -r requirements.txt
3. Configure environment
# Copy the example environment file
cp .env.example .env
# IMPORTANT: Set secure permissions (Unix/Linux/macOS)
chmod 600 .env
# Edit .env with your n8n credentials
4. Configure Claude Code
Add to your Claude Code settings (~/.claude/settings.json):
Windows:
{
"mcpServers": {
"n8n": {
"command": "py",
"args": ["-3", "C:\\path\\to\\n8n-mcp-server\\server.py"]
}
}
}
macOS/Linux:
{
"mcpServers": {
"n8n": {
"command": "python3",
"args": ["/path/to/n8n-mcp-server/server.py"]
}
}
}
Configuration
Required Variables
| Variable | Description | Example |
|---|---|---|
N8N_URL |
n8n instance URL (HTTPS recommended) | https://n8n.example.com |
N8N_API_KEY |
n8n API key (min 32 chars) | eyJhbGciOiJIUzI1... |
Security Variables
| Variable | Description | Default |
|---|---|---|
N8N_REQUIRE_HTTPS |
Require HTTPS for non-localhost | true |
N8N_VERIFY_TLS |
Verify TLS certificates | true |
AUDIT_LOG_ENABLED |
Enable audit logging | true |
Optional Variables
| Variable | Description | Default |
|---|---|---|
N8N_REQUEST_TIMEOUT |
Request timeout (seconds) | 30 |
N8N_MAX_PAYLOAD_SIZE |
Max payload size (bytes) | 10485760 |
LOG_LEVEL |
Logging level | INFO |
Available Tools
Workflow Management
| Tool | Description |
|---|---|
n8n_list_workflows |
List all workflows with optional filtering |
n8n_get_workflow |
Get details of a specific workflow |
n8n_create_workflow |
Create a new workflow |
n8n_update_workflow |
Update an existing workflow |
n8n_delete_workflow |
Delete a workflow |
n8n_activate_workflow |
Activate a workflow (enable triggers) |
n8n_deactivate_workflow |
Deactivate a workflow (disable triggers) |
n8n_import_workflow |
Import a workflow from JSON |
Execution Management
| Tool | Description |
|---|---|
n8n_execute_workflow |
Manually execute a workflow |
n8n_list_executions |
List execution history with filters |
n8n_get_execution |
Get details of a specific execution |
Other
| Tool | Description |
|---|---|
n8n_list_credentials |
List available credentials (metadata only) |
n8n_health_check |
Check n8n connectivity and health |
Example Usage
List all workflows:
"Show me all active workflows in n8n"
Create a workflow:
"Create a workflow that triggers daily at 8am and sends a Slack message"
Debug a failed execution:
"Why did my last workflow execution fail?"
Health check:
"Check if n8n is connected and working"
Audit Logging
When AUDIT_LOG_ENABLED=true, all operations are logged in JSON format:
{
"timestamp": "2024-01-15T10:30:00.000Z",
"correlation_id": "a1b2c3d4e5f6",
"operation": "create_workflow",
"success": true,
"duration_ms": 125.5,
"parameters": {"name": "My Workflow"}
}
Project Structure
n8n-mcp-server/
├── server.py # Main MCP server (enterprise security)
├── requirements.txt # Python dependencies
├── .env.example # Configuration template
├── .env # Your configuration (git-ignored)
├── .gitignore # Git ignore rules
├── SECURITY.md # Security policy and documentation
├── SPEC.md # API specification
└── README.md # This file
Troubleshooting
"HTTPS required" error
- Your n8n URL uses HTTP instead of HTTPS
- For development only: set
N8N_REQUIRE_HTTPS=false
"TLS verification failed"
- Certificate issue with your n8n instance
- For self-signed certs in development: set
N8N_VERIFY_TLS=false
"Rate limit exceeded"
- Too many requests in a short time
- Wait and retry, or check for loops in your usage
"Authentication failed"
- Check
N8N_API_KEYis correct - Verify API key hasn't expired
- Ensure API key has required permissions
Development
# Run with debug logging
LOG_LEVEL=DEBUG python server.py
# Run security scan
pip install bandit
bandit -r server.py
# Check dependencies for vulnerabilities
pip install safety
safety check -r requirements.txt
Contributing
- Fork the repository
- Create a feature branch
- Make your changes (security-impacting changes require review)
- Submit a pull request
See SECURITY.md for security guidelines.
License
MIT License - see LICENSE for details.
Changelog
v2.0.0 (2024) - Enterprise Security Edition
- TLS/SSL enforcement by default
- Comprehensive audit logging with correlation IDs
- Rate limiting protection
- Enhanced input validation with injection detection
- Sensitive data redaction
- Security policy documentation
- .env file permission checking
v1.2.0
- Added
.envfile support - Added health check tool
- Improved logging
v1.0.0
- Initial release
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。