n8n MCP Bridge

MCP Tools

16 powerful tools for complete n8n automation

MCP Tools

n8n MCP Bridge provides 16 comprehensive tools that enable AI assistants to fully automate your n8n workflows.

What are MCP Tools?

MCP (Model Context Protocol) tools are functions that AI assistants can call to perform specific actions. Each tool has:

  • Name: Unique identifier (e.g., n8n_list_workflows)
  • Description: What the tool does
  • Parameters: Input required
  • Returns: Output provided

Available Tools (16)

System Tools (2)

ping_n8n

Test connectivity to your n8n instance and verify API key.

Usage: Can you ping my n8n instance?

View Details →

n8n_health_check

Comprehensive health check of n8n instance, database, and MCP server.

Modes: quick (fast) or diagnostic (detailed)

Usage: Run a health check on my n8n setup


Workflow Management (5)

n8n_list_workflows

List all workflows with filtering, pagination, and sorting.

Parameters:

  • active (boolean, optional): Filter by active status
  • tags (array, optional): Filter by tags
  • limit (number, optional): Max results (default: 100, max: 500)
  • sortBy (string, optional): Sort field (name, createdAt, updatedAt)

Usage: Show me all active workflows tagged "production"

n8n_get_workflow

Get detailed information about a specific workflow including all nodes and connections.

Parameters:

  • workflowId (string, required): Workflow ID

Usage: Show me the details of workflow abc123

n8n_create_workflow

Create a new workflow in your n8n instance (created as INACTIVE for safety).

Parameters:

  • name (string, required): Workflow name
  • nodes (array, required): Array of workflow nodes
  • connections (object, required): Node connections
  • settings (object, optional): Workflow settings

Usage: Create a workflow that sends a Slack message when a webhook is received

n8n_update_workflow

Update existing workflows (metadata or full structure).

Parameters:

  • workflowId (string, required): Workflow ID
  • name (string, optional): New name
  • tags (array, optional): New tags
  • nodes (array, optional): Updated nodes (INACTIVE only)
  • connections (object, optional): Updated connections (INACTIVE only)

Usage: Rename workflow abc123 to "Production Newsletter"

n8n_delete_workflow

Delete a workflow from your n8n instance (requires confirmation).

Parameters:

  • workflowId (string, required): Workflow ID
  • confirm (boolean, required): Must be true to confirm deletion

Usage: Delete workflow abc123 (will ask for confirmation)


Execution History (2)

n8n_list_executions

List recent workflow executions with status and filtering.

Parameters:

  • workflowId (string, optional): Filter by workflow
  • status (string, optional): Filter by status (success, error, running, waiting, canceled)
  • limit (number, optional): Max results (default: 20, max: 100)

Usage: Show me the last 10 failed executions

n8n_get_execution

Get detailed execution data including all node outputs and errors.

Parameters:

  • executionId (string, required): Execution ID

Usage: Show me the error details for execution exec_123


Node Discovery (2)

n8n_search_nodes

Search through 525+ n8n nodes with full-text search.

Parameters:

  • query (string, required): Search term
  • package (string, optional): Filter by package (base, langchain)
  • limit (number, optional): Max results (default: 50, max: 100)

Usage: Find nodes related to email

n8n_get_node_details

Get comprehensive details about a specific n8n node.

Parameters:

  • nodeId (string, required): Node ID (e.g., "nodes-base.slack")
  • detailed (boolean, optional): Return full schema (default: false)

Usage: Tell me about the Slack node


Tag Management (5) ✨ NEW

n8n_list_tags

List all tags in your n8n instance.

Usage: Show me all my workflow tags

n8n_create_tag

Create a new tag for workflow organization.

Parameters:

  • name (string, required): Tag name (1-50 characters)

Usage: Create a tag called "production"

n8n_delete_tag

Delete a tag (removes it from all workflows, requires confirmation).

Parameters:

  • tagId (string, required): Tag ID
  • confirm (boolean, required): Must be true to confirm deletion

Usage: Delete tag tag_123

n8n_assign_tags_to_workflow

Assign tags to a workflow (auto-creates missing tags).

Parameters:

  • workflowId (string, required): Workflow ID
  • tags (array, required): Array of tag names
  • createIfMissing (boolean, optional): Auto-create tags (default: true)

Usage: Tag workflow abc123 with "production" and "scheduled"

n8n_remove_tags_from_workflow

Remove tags from a specific workflow.

Parameters:

  • workflowId (string, required): Workflow ID
  • tagIds (array, required): Array of tag IDs to remove

Usage: Remove tag tag_123 from workflow abc123


Tool Categories

Production Tools ✅

All 16 tools are production-ready and available now:

  • ✅ 2 System tools
  • ✅ 5 Workflow management tools
  • ✅ 2 Execution history tools
  • ✅ 2 Node discovery tools
  • ✅ 5 Tag management tools

Security Features

  • 🔒 Multi-tenant isolation - Each API key only accesses its own data
  • 🔒 Encrypted credentials - n8n API keys encrypted with AES-256
  • 🔒 Confirmation required - Destructive actions require explicit confirm=true
  • 🔒 Safety defaults - New workflows created as INACTIVE

Using Tools

Via Claude Desktop

Ask Claude naturally:

Show me all active workflows tagged "production"

Claude will automatically use the n8n_list_workflows tool.

Via Claude Code

Can you create a workflow that sends a Slack message?

Claude will use n8n_create_workflow and n8n_search_nodes tools.

Via API

Call tools directly via MCP protocol:

{
  "method": "tools/call",
  "params": {
    "name": "n8n_list_workflows",
    "arguments": {
      "active": true,
      "tags": ["production"]
    }
  }
}

Tool Responses

All tools return standardized responses:

Success Response

{
  "success": true,
  "data": { ... },
  "suggestions": ["Next steps..."],
  "metadata": { ... }
}

Error Response

{
  "success": false,
  "error": "Error message",
  "suggestions": ["How to fix..."]
}

Memory Integration

Tools can access memory items for context:

Memory Item:

Label: Production Workflow Rules
Content: Always test workflows in dev before deploying to production
Applies To: n8n_update_workflow, n8n_create_workflow

Tool Behavior:

User: "Create a payment workflow"
AI: "I'll create the payment workflow. Based on your memory,
should I create it as INACTIVE for testing first?"

Learn About Memory System →

Analytics

View tool usage in your dashboard:

  • Total Calls: How many times each tool was called
  • Success Rate: Percentage of successful calls
  • Popular Tools: Most frequently used
  • Error Tracking: Failed calls and reasons

Best Practices

Tool Selection

Choose the right tool for the job:

  • n8n_health_check - Test connectivity before starting
  • n8n_list_workflows - Discover available workflows
  • n8n_get_workflow - View workflow before modifying
  • n8n_create_workflow - Build new automations
  • n8n_assign_tags_to_workflow - Organize workflows

Workflow Safety

  • Always use n8n_get_workflow to review before updating
  • New workflows are created as INACTIVE (safe)
  • Test workflows manually in n8n UI before activating
  • Use tags to organize production vs development workflows

Tag Organization

  • Use consistent naming: production, dev, scheduled, webhook
  • Tag workflows by environment, purpose, or team
  • Use n8n_list_workflows with tag filters to find workflows
  • Clean up unused tags with n8n_delete_tag

Limitations

Current limitations:

  • No workflow execution - Security decision (execute manually in n8n UI)
  • No batch operations - One tool call at a time
  • No streaming - Results returned after completion

These are intentional design decisions for security and reliability.

Troubleshooting

Tool Not Found

If AI can't find a tool:

  1. Verify API key is valid (ping_n8n)
  2. Check MCP server is running (n8n_health_check)
  3. Ensure you're using correct tool name
  4. Reconnect MCP client (restart Claude Desktop)

Tool Execution Failed

If tool fails:

  1. Check parameters are correct
  2. Verify n8n connection (ping_n8n)
  3. Review error message details
  4. Check n8n API key permissions

Slow Tool Response

If tool is slow:

  1. Check n8n instance performance
  2. Verify network connectivity
  3. Use pagination for large result sets
  4. Run n8n_health_check diagnostic mode

Next Steps

On this page