What are the system requirements?
Node.js v18.0.0 or higher, a running n8n instance (self-hosted or cloud), and an n8n API key with appropriate permissions.
@makafeli
Basic information
AI assistant integration for n8n workflow automation through Model Context Protocol (MCP). Connect Claude Desktop, ChatGPT, and other AI assistants to n8n for natural language workflow management.
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"n8n-workflow-builder": {
"command": "npx",
"args": [
"@makafeli/n8n-workflow-builder"
]
}
}
}List all workflows from n8n instance
Create a new workflow in n8n
Get a workflow by ID
Update an existing workflow by ID
Delete a workflow by ID
Activate a workflow by ID
Deactivate a workflow by ID
Execute a workflow manually
Create a new workflow and immediately activate it
List workflow executions with filtering and pagination support
Get detailed information about a specific workflow execution
Delete a workflow execution record from the n8n instance
List all workflow tags with pagination support
Create a new workflow tag for organization and categorization
Retrieve individual tag details by ID
Modify tag names for better organization
Remove unused tags from the system
Get all tags associated with a specific workflow
Assign or remove tags from workflows
Create a new credential for workflow authentication. Use get_credential_schema first to understand required fields for the credential type.
Get the schema for a specific credential type to understand what fields are required when creating credentials.
Delete a credential by ID. This will remove the credential and make it unavailable for workflows. Use with caution as this action cannot be undone.
Generate a comprehensive security audit report for the n8n instance
A Model Context Protocol (MCP) server that bridges AI assistants (such as Claude Desktop, ChatGPT, or any MCP-compatible client) with an n8n automation instance. It enables natural-language workflow management, creation, execution, and monitoring through the n8n API.
Install via NPX (npx @makafeli/n8n-workflow-builder) or use the hosted Smithery.ai version. Set environment variables N8N_HOST (your n8n URL) and N8N_API_KEY (from n8n Settings > API Keys). Configure the server in your MCP client (e.g., Claude Desktop, Cline) with the same command and env values.
Node.js v18.0.0 or higher, a running n8n instance (self-hosted or cloud), and an n8n API key with appropriate permissions.
In your n8n instance, go to Settings → API Keys → Create API Key, then copy the generated key.
Yes, the server is available as a hosted service on Smithery.ai with no local installation required.
The server works with any MCP-compatible client; explicit setup instructions are provided for Claude Desktop and Cline (VS Code extension).
Set `N8N_HOST` to your n8n instance URL (e.g., `http://localhost:5678`) and `N8N_API_KEY` to your API key, then export them or include them in your MCP client’s config JSON.
MCP server that gives Claude persistent project memory — roadmap, session history, and codebase context. Type continue and Claude picks up exactly where you left off
This project demonstrates a decoupled real-time agent architecture that connects LangGraph agents to remote tools served by custom MCP (Modular Command Protocol) servers. The architecture enables a flexible and scalable multi-agent system where each tool can be hosted independent
Neuronto Agentic Resource Discovery (ARD) Index. Federated search across every public ARD registry, plus a verified tool index read from each MCP server's own tools/list, hybrid lexical and semantic retrieval, and ARD-Bench.
This project demonstrates a decoupled real-time agent architecture that connects LangGraph agents to remote tools served by custom MCP (Modular Command Protocol) servers. The architecture enables a flexible and scalable multi-agent system where each tool can be hosted independent
MCP server that allow LLM in agent mode builds n8n workflows for you
Comments