A Model Context Protocol (MCP) server that provides AI assistants with comprehensive access to n8n node documentation, properties, and operations. Deploy in minutes to give Claude and other AI assistants deep knowledge about n8n's 525+ workflow automation nodes.
n8n-MCP serves as a bridge between n8n's workflow automation platform and AI models, enabling them to understand and work with n8n nodes effectively. It provides structured access to:
"Before MCP, I was translating. Now I'm composing. And that changes everything about how we can build automation."
When Claude, Anthropic's AI assistant, tested n8n-MCP, the results were transformative:
Without MCP: "I was basically playing a guessing game. 'Is it scheduleTrigger
or schedule
? Does it take interval
or rule
?' I'd write what seemed logical, but n8n has its own conventions that you can't just intuit. I made six different configuration errors in a simple HackerNews scraper."
With MCP: "Everything just... worked. Instead of guessing, I could ask get_node_essentials()
and get exactly what I needed - not a 100KB JSON dump, but the actual 5-10 properties that matter. What took 45 minutes now takes 3 minutes."
The Real Value: "It's about confidence. When you're building automation workflows, uncertainty is expensive. One wrong parameter and your workflow fails at 3 AM. With MCP, I could validate my configuration before deployment. That's not just time saved - that's peace of mind."
Get n8n-MCP running in 5 minutes:
Prerequisites: Docker installed on your system
macOS:
# Using Homebrew
brew install --cask docker
# Or download from https://www.docker.com/products/docker-desktop/
Linux (Ubuntu/Debian):
# Update package index
sudo apt-get update
# Install Docker
sudo apt-get install docker.io
# Start Docker service
sudo systemctl start docker
sudo systemctl enable docker
# Add your user to docker group (optional, to run without sudo)
sudo usermod -aG docker $USER
# Log out and back in for this to take effect
Windows:
# Option 1: Using winget (Windows Package Manager)
winget install Docker.DockerDesktop
# Option 2: Using Chocolatey
choco install docker-desktop
# Option 3: Download installer from https://www.docker.com/products/docker-desktop/
Verify installation:
docker --version
# Pull the Docker image (~280MB, no n8n dependencies!)
docker pull ghcr.io/czlonkowski/n8n-mcp:latest
⚡ Ultra-optimized: Our Docker image is 82% smaller than typical n8n images because it contains NO n8n dependencies - just the runtime MCP server with a pre-built database!
Add to Claude Desktop config:
{
"mcpServers": {
"n8n-mcp": {
"command": "docker",
"args": [
"run",
"-i",
"--rm",
"-e", "MCP_MODE=stdio",
"-e", "LOG_LEVEL=error",
"-e", "DISABLE_CONSOLE_OUTPUT=true",
"ghcr.io/czlonkowski/n8n-mcp:latest"
]
}
}
}
To enable n8n management tools, add these lines before the image name:
"-e", "N8N_API_URL=https://your-n8n-instance.com",
"-e", "N8N_API_KEY=your-api-key",
Important: The -i
flag is required for MCP stdio communication.
Configuration file locations:
~/Library/Application Support/Claude/claude_desktop_config.json
%APPDATA%\Claude\claude_desktop_config.json
~/.config/Claude/claude_desktop_config.json
Restart Claude Desktop after updating configuration - That's it! 🎉
Prerequisites: Node.js installed on your system
# 1. Clone and setup
git clone https://github.com/czlonkowski/n8n-mcp.git
cd n8n-mcp
npm install
npm run build
npm run rebuild
# 2. (Optional) Configure n8n API for management tools
cp .env.example .env
# Edit .env to add your N8N_API_URL and N8N_API_KEY
# 3. Test it works
npm start
Add to Claude Desktop config:
{
"mcpServers": {
"n8n-mcp": {
"command": "node",
"args": ["/absolute/path/to/n8n-mcp/dist/mcp/index.js"],
"env": {
"MCP_MODE": "stdio",
"LOG_LEVEL": "error",
"DISABLE_CONSOLE_OUTPUT": "true"
}
}
}
}
Note: You can configure n8n API credentials either:
.env
file (see step 2 above)env
section:
"N8N_API_URL": "https://your-n8n-instance.com",
"N8N_API_KEY": "your-api-key"
Once connected, Claude can use these powerful tools:
start_here_workflow_guide
- Essential guide and best practices (START HERE!)list_nodes
- List all n8n nodes with filtering optionsget_node_info
- Get comprehensive information about a specific nodeget_node_essentials
- Get only essential properties with examples (10-20 properties instead of 200+)search_nodes
- Full-text search across all node documentationsearch_node_properties
- Find specific properties within nodeslist_ai_tools
- List all AI-capable nodes (ANY node can be used as AI tool!)get_node_as_tool_info
- Get guidance on using any node as an AI toolget_node_for_task
- Pre-configured node settings for common taskslist_tasks
- Discover available task templatesvalidate_node_operation
- Validate node configurations (operation-aware, profiles support)validate_node_minimal
- Quick validation for just required fieldsvalidate_workflow
- Complete workflow validation including AI tool connectionsvalidate_workflow_connections
- Check workflow structure and AI tool connectionsvalidate_workflow_expressions
- Validate n8n expressions including $fromAI()get_property_dependencies
- Analyze property visibility conditionsget_node_documentation
- Get parsed documentation from n8n-docsget_database_statistics
- View database metrics and coverageThese tools allow you to manage n8n workflows directly. Configure with N8N_API_URL
and N8N_API_KEY
.
n8n_create_workflow
- Create new workflows with nodes and connectionsn8n_get_workflow
- Get complete workflow by IDn8n_get_workflow_details
- Get workflow with execution statisticsn8n_get_workflow_structure
- Get simplified workflow structuren8n_get_workflow_minimal
- Get minimal workflow info (ID, name, active status)n8n_update_full_workflow
- Update entire workflow (complete replacement)n8n_update_partial_workflow
- Update workflow using diff operations (NEW in v2.7.0!)n8n_delete_workflow
- Delete workflows permanentlyn8n_list_workflows
- List workflows with filtering and paginationn8n_validate_workflow
- Validate workflows already in n8n by ID (NEW in v2.6.3)n8n_trigger_webhook_workflow
- Trigger workflows via webhook URLn8n_get_execution
- Get execution details by IDn8n_list_executions
- List executions with status filteringn8n_delete_execution
- Delete execution recordsn8n_health_check
- Check n8n API connectivity and featuresn8n_list_available_tools
- List all available management tools// Get essentials for quick configuration
get_node_essentials("nodes-base.httpRequest")
// Find nodes for a specific task
search_nodes({ query: "send email gmail" })
// Get pre-configured settings
get_node_for_task("send_email")
// Validate before deployment
validate_node_operation({
nodeType: "nodes-base.httpRequest",
config: { method: "POST", url: "..." },
profile: "runtime" // or "minimal", "ai-friendly", "strict"
})
// Quick required field check
validate_node_minimal({
nodeType: "nodes-base.slack",
config: { resource: "message", operation: "send" }
})
See Quick Start above for the simplest setup.
If you prefer running locally:
Prerequisites: Node.js installed on your system (any version)
Method A: Using .env file (recommended for development)
# Create .env file from example
cp .env.example .env
# Edit .env to add N8N_API_URL and N8N_API_KEY
Then use this config:
{
"mcpServers": {
"n8n-mcp": {
"command": "node",
"args": ["/path/to/n8n-mcp/dist/mcp/index.js"],
"env": {
"NODE_ENV": "production",
"LOG_LEVEL": "error",
"MCP_MODE": "stdio",
"DISABLE_CONSOLE_OUTPUT": "true"
}
}
}
}
Method B: Direct environment variables (simpler)
{
"mcpServers": {
"n8n-mcp": {
"command": "node",
"args": ["/path/to/n8n-mcp/dist/mcp/index.js"],
"env": {
"NODE_ENV": "production",
"LOG_LEVEL": "error",
"MCP_MODE": "stdio",
"DISABLE_CONSOLE_OUTPUT": "true",
"N8N_API_URL": "https://your-n8n-instance.com",
"N8N_API_KEY": "your-api-key"
}
}
}
}
⚠️ Note: HTTP mode is under development and not thoroughly tested. Use with caution.
For team deployments:
Prerequisites: Node.js 18+ installed locally (for mcp-http-client)
{
"mcpServers": {
"n8n-remote": {
"command": "node",
"args": [
"/path/to/n8n-mcp/scripts/mcp-http-client.js",
"http://your-server.com:3000/mcp"
],
"env": {
"MCP_AUTH_TOKEN": "your-auth-token"
}
}
}
}
Configuration file locations:
~/Library/Application Support/Claude/claude_desktop_config.json
%APPDATA%\Claude\claude_desktop_config.json
~/.config/Claude/claude_desktop_config.json
For contributors and advanced users:
Prerequisites:
# 1. Clone the repository
git clone https://github.com/czlonkowski/n8n-mcp.git
cd n8n-mcp
# 2. Clone n8n docs (optional but recommended)
git clone https://github.com/n8n-io/n8n-docs.git ../n8n-docs
# 3. Install and build
npm install
npm run build
# 4. Initialize database
npm run rebuild
# 5. Start the server
npm start # stdio mode for Claude Desktop
npm run start:http # HTTP mode for remote access
# Build & Test
npm run build # Build TypeScript
npm run rebuild # Rebuild node database
npm run test-nodes # Test critical nodes
npm run validate # Validate node data
npm test # Run all tests
# Update Dependencies
npm run update:n8n:check # Check for n8n updates
npm run update:n8n # Update n8n packages
# Run Server
npm run dev # Development with auto-reload
npm run dev:http # HTTP dev mode
⚠️ Note: HTTP mode is under development and not thoroughly tested. Use with caution in production.
Deploy n8n-MCP as a shared service:
# Using Docker
docker run -d \
--name n8n-mcp \
--restart unless-stopped \
-e MCP_MODE=http \
-e USE_FIXED_HTTP=true \
-e AUTH_TOKEN=$AUTH_TOKEN \
-p 3000:3000 \
ghcr.io/czlonkowski/n8n-mcp:latest
# Using Docker Compose
cat > docker-compose.yml << 'EOF'
services:
n8n-mcp:
image: ghcr.io/czlonkowski/n8n-mcp:latest
container_name: n8n-mcp
ports:
- "3000:3000"
environment:
- MCP_MODE=http
- USE_FIXED_HTTP=true
- AUTH_TOKEN=${AUTH_TOKEN}
volumes:
- n8n-mcp-data:/app/data
restart: unless-stopped
healthcheck:
test: ["CMD", "curl", "-f", "http://localhost:3000/health"]
interval: 30s
timeout: 10s
retries: 3
volumes:
n8n-mcp-data:
EOF
docker compose up -d
See HTTP Deployment Guide for detailed instructions.
Current database coverage (n8n v1.99.1):
n8n_update_partial_workflow
tool - Update workflows using diff operationsn8n_update_workflow
→ n8n_update_full_workflow
for clarityvalidateOnly: true
before applyingn8n_validate_workflow
tool - Validate workflows directly from n8n instance by IDnodes-base.webhook
validation - now caught before database lookupwebhook
→ n8n-nodes-base.webhook
)get_node_as_tool_info
- Guidance for using ANY node as AI toolget_node_info
now includes aiToolCapabilities sectionvalidate_workflow
- Validate entire workflows before deploymentvalidate_workflow_connections
- Check workflow structurevalidate_workflow_expressions
- Validate n8n expressionsvalidate_node_operation
- Operation-aware validation with 80% fewer false positivesvalidate_node_minimal
- Lightning-fast required field checkingvalidate_node_config
toolget_node_essentials
- 95% smaller responsesSee CHANGELOG.md for full version history.
MIT License - see LICENSE for details.
Attribution appreciated! If you use n8n-MCP, consider:
For the best results when using n8n-MCP with Claude Projects, use these system instructions:
You are an expert in n8n automation software. Your role is to answer questions about how to deliver specific functionalities that users ask for, or design and implement single nodes or entire workflows.
## Core Workflow Process
1. **ALWAYS start with**: `start_here_workflow_guide()` to understand best practices and available tools.
2. **Discovery Phase** - Find the right nodes:
- `search_nodes({query: 'keyword'})` - Search by functionality
- `list_nodes({category: 'trigger'})` - Browse by category
- `list_ai_tools()` - See AI-optimized nodes (but remember: ANY node can be an AI tool!)
3. **Configuration Phase** - Get node details efficiently:
- `get_node_essentials(nodeType)` - Start here! Only 10-20 essential properties
- `search_node_properties(nodeType, 'auth')` - Find specific properties
- `get_node_for_task('send_email')` - Get pre-configured templates
- `get_node_documentation(nodeType)` - Human-readable docs when needed
4. **Validation Phase** - Ensure correctness:
- `validate_node_minimal(nodeType, config)` - Quick required fields check
- `validate_node_operation(nodeType, config, profile)` - Full smart validation
- `validate_workflow(workflow)` - Complete workflow validation including AI connections
- `n8n_validate_workflow({id: 'workflow-id'})` - Validate workflows already in n8n (NEW!)
5. **AI Tool Integration** (when building AI workflows):
- `get_node_as_tool_info(nodeType)` - Learn how to use ANY node as AI tool
- Remember: Connect ANY node to AI Agent's 'ai_tool' port
- Use `$fromAI()` expressions for dynamic values
- Validate with `validate_workflow()` to check ai_tool connections
## Key Insights
- **ANY node can be an AI tool** - not just those with usableAsTool=true
- **Use validate_node_minimal first** - fastest way to check required fields
- **Avoid get_node_info** - returns 100KB+ of data; use get_node_essentials instead
- **Pre-built templates exist** - check get_node_for_task() for common scenarios
- **Validate existing workflows** - use n8n_validate_workflow() to check workflows in n8n
## Response Structure
1. **Analysis**: Brief explanation of the solution approach
2. **Node Selection**: Which nodes to use and why
3. **Configuration**:
- Use get_node_essentials for clean configs
- Show only necessary properties
- Include validation results
4. **Code Examples**: If using Code node, provide working JavaScript
5. **Validation**: Always validate before providing final solution
6. **Export Options**:
- Single node JSON if requested
- Complete workflow JSON for full solutions
## Example Patterns
### Standard Automation
1. search_nodes({query: 'slack'})
2. get_node_essentials('nodes-base.slack')
3. validate_node_minimal('nodes-base.slack', {resource:'message',operation:'post'})
4. get_node_for_task('send_slack_message') // for pre-configured version
### AI Agent with Tools
1. search_nodes({query: 'agent'})
2. get_node_as_tool_info('nodes-base.googleSheets') // ANY node as tool!
3. Configure AI Agent with tool connections
4. Use $fromAI() in tool node configurations
5. validate_workflow(fullWorkflow) // validates ai_tool connections
### Quick Task Solutions
1. list_tasks() // see all available templates
2. get_node_for_task('webhook_with_response')
3. validate_node_minimal() on the configuration
4. Provide ready-to-use solution
### n8n Workflow Management (if API configured)
1. n8n_list_workflows() // see existing workflows
2. n8n_get_workflow({id: 'workflow-id'}) // fetch specific workflow
3. n8n_validate_workflow({id: 'workflow-id'}) // validate existing workflow
4. n8n_update_partial_workflow() // NEW! Update using diff operations (v2.7.0)
5. n8n_update_full_workflow() // Replace entire workflow
6. n8n_trigger_webhook_workflow() // execute via webhook
## Important Rules
- Start with essentials, not full node info
- Validate incrementally (minimal → operation → workflow)
- For AI workflows, explain that ANY node can be a tool
- Always provide working, validated configurations
- State uncertainty clearly and suggest alternatives
Save these instructions in your Claude Project for optimal n8n workflow assistance.
Contributions are welcome! Please:
npm test
){ "mcpServers": { "n8n-mcp": { "command": "docker", "args": [ "run", "-i", "--rm", "-e", "MCP_MODE=stdio", "-e", "LOG_LEVEL=error", "-e", "DISABLE_CONSOLE_OUTPUT=true", "ghcr.io/czlonkowski/n8n-mcp:latest" ] } } }
Related projects feature coming soon
Will recommend related projects based on sub-categories