Build your own N8N Workflows MCP Server
VerifiedExpose n8n workflows as an MCP server for AI agents.
What this workflow does
This workflow converts existing n8n automations into discoverable MCP server endpoints that agents can list, manage, and invoke through structured tool calls.
It is intended for AI developers and power users who want agents to execute complete n8n workflows instead of isolated actions.
Who is this for?
n8n power users and automation engineers who already maintain complex workflows and want to expose them to MCP clients like Claude Desktop. AI agent builders and technical teams integrating outcome-focused tools rather than granular API calls.
What problem it solves
MCP clients are limited to simple tools and require many sequential calls to accomplish real tasks. This forces agents to orchestrate low-level steps instead of triggering complete, reliable end-to-end n8n workflows.
What it automates
Expose internal automation library to Claude
Connect Claude Desktop to an existing set of n8n workflows so the agent can discover and run multi-step processes like lead enrichment or report generation in one call.
Reduce agent round-trips for complex tasks
Let an AI agent trigger a single n8n workflow that handles data fetching, transformation, and notification instead of making separate tool calls for each step.
Central management of workflow permissions
Use the MCP server layer to control which workflows are visible and executable by different MCP clients without modifying the original workflows.
How the workflow works
The 6 nodes in this automation, in order.
- 1Redisredis
- 2AI Agent@n8n/n8n-nodes-langchain.agent
- 3OpenAI Chat Model@n8n/n8n-nodes-langchain.lmChatOpenAi
- 4Simple Memory@n8n/n8n-nodes-langchain.memoryBufferWindow
- 5Call n8n Workflow Tool@n8n/n8n-nodes-langchain.toolWorkflow
- 6MCP Client Tool@n8n/n8n-nodes-langchain.mcpClientTool
Apps & integrations used
How to set up Build your own N8N Workflows MCP Server
- 1Import the template into your n8n instance
- 2Configure the MCP Client Tool node with your client credentials
- 3Connect the AI Agent to the OpenAI Chat Model and Simple Memory nodes
- 4Add your existing workflows as targets for the Call n8n Workflow Tool
- 5Set up Redis for workflow discovery and state tracking
- 6Activate the workflow and test discovery from your MCP client
How to customize this workflow
- →Swap OpenAI Chat Model for another supported LLM provider
- →Add extra Call n8n Workflow Tool nodes for additional workflow groups
- →Change the trigger from MCP Client Tool to a different MCP-compatible trigger
- →Extend Simple Memory with Redis keys for longer conversation context
Build your own N8N Workflows MCP Server: pros & cons
Pros
- +Turns existing n8n workflows into powerful MCP tools
- +Reduces number of calls an agent must make
- +Works with any MCP client (Claude Desktop, etc.)
- +Leverages n8n's visual workflow builder for complex logic
Cons
- –Requires an active n8n instance and Redis
- –Intermediate setup complexity for non-technical users
- –Discovery only works for workflows explicitly exposed
Frequently asked questions
It turns your n8n workflows into an MCP server so MCP clients can discover and run them as single high-level tools.
User reviews
Verified reviews from the community shape this listing's rating.
Loading reviews…
Similar workflows
Other ai & llm automations worth a look.