n8n-mcp

MCP server for n8n — gives Claude, Cursor, and other AI agents tools for generating workflows, linting, diagnosing failed executions, and driving live n8n…

CommunityOnlineNo sign-inGlobalFreeRead-only

What it can do

  • Node.Scaffold: Scaffold a TypeScript skeleton for an n8n custom node from a plain-English description. Returns a single TypeScript file implementing INodeType with description, credentials reference,
  • Workflow.Generate: Generate a valid n8n workflow JSON from a plain-English description. Handles webhook/schedule/RSS triggers, common action nodes (Slack, Google Sheets, Discord, Gmail, Notion, HTTP),
  • Workflow.Lint: Lint an n8n workflow JSON. Returns concrete errors and warnings: missing credentials, deprecated node types (Function -> Code, spreadsheetFile -> convertToFile/extractFromFile), broken

What data it sees

Do you need an account

No: the server works without sign-in

MCP server for n8n — gives Claude, Cursor, and other AI agents tools for generating workflows, linting, diagnosing failed executions, and driving live n8n instances.

How to connect

How to connect n8n-mcp to Claude

The server address to paste into Claude settings:

https://n8n-mcp--automatelab.run.tools

  1. Open Claude (claude.ai or the desktop app).
  2. Go to Settings → Connectors.
  3. Click Add custom connector.
  4. Paste the server address copied below into Remote MCP server URL and click Add.
  5. In a chat, click + → Connectors and switch the new connector on.

Custom connectors are available on Free, Pro, Max, Team and Enterprise plans (Free is limited to one). On Team and Enterprise an organization Owner adds the connector first under Organization settings → Connectors.

Authorization

After you click Add, a sign-in window for the service opens. Sign in with your own account and approve access. Claude never sees your password.

How to connect n8n-mcp to ChatGPT

The server address to paste into ChatGPT settings:

https://n8n-mcp--automatelab.run.tools

  1. Open ChatGPT in a browser (chatgpt.com). A Plus, Pro, Business, Enterprise or Edu plan is required.
  2. Turn on developer mode once: Settings → Apps → Advanced settings → Developer mode.
  3. Open Settings → Connectors and click Create.
  4. Fill in the form: Name (anything), Description (one line about what the service does), MCP server URL (copy it below).
  5. Under Authentication choose OAuth if the service requires sign-in, otherwise None. Click Create.
  6. In a new chat open + → Apps/Connectors and enable the connector.

OpenAI has renamed this section before (Connectors → Apps/Plugins). If the label differs, search settings for "developer mode". On Business/Enterprise workspaces an admin must allow custom connectors first.

Authorization

On first use ChatGPT opens the service's sign-in window. Sign in and approve access.

How to connect n8n-mcp to Cursor

The server address to paste into Cursor settings:

https://n8n-mcp--automatelab.run.tools

Fastest: click Open in Cursor below and confirm the prompt.

Manually:

  1. In Cursor open Settings → Cursor Settings → MCP and click Add new global MCP server.
  2. Paste the JSON copied below into ~/.cursor/mcp.json (per project: .cursor/mcp.json in the repo root).
  3. Save the file. The server appears in the MCP list; authorize it there if asked.

Authorization

If the service needs sign-in, an authorize button appears next to the server in the MCP list.

Server tool list (12)

Raw names from tools/list. Only developers need these.

node.scaffoldScaffold a TypeScript skeleton for an n8n custom node from a plain-English description. Returns a single TypeScript file implementing INodeType with description, credentials reference, and an execute method stub. Pure code generation — no network, no filesystem writes.
workflow.generateGenerate a valid n8n workflow JSON from a plain-English description. Handles webhook/schedule/RSS triggers, common action nodes (Slack, Google Sheets, Discord, Gmail, Notion, HTTP), and AI Agent setups (LangChain root agent + chat model + memory + optional HTTP tool, wired with ai_languageModel / ai_memory / ai_tool connections). Returns workflow JSON with unique node IDs, connections, positions, and typeVersion on every node. Output is non-deterministic (random node IDs and webhook paths).
workflow.lintLint an n8n workflow JSON. Returns concrete errors and warnings: missing credentials, deprecated node types (Function -> Code, spreadsheetFile -> convertToFile/extractFromFile), broken connections, missing or non-numeric typeVersion, duplicate node names or IDs, AI Agent missing ai_languageModel sub-node, Webhook missing webhookId, IF node still on v1 condition schema, rate-sensitive nodes without retries, Code-node sandbox violations, expression staleness (`$('NodeName')` referencing missing nodes), manualTrigger in active workflows, disabled-but-wired nodes, empty Set nodes, HTTP method/body mismatches, Schedule trigger DST risk, credential drift, webhook test paths in active workflows. Deterministic, rule-based.
workflow.diffSemantic diff between two workflows. Reports nodes added / removed / modified (with field-level deltas: type, typeVersion, parameters, credentials, disabled, position), connection topology changes, and settings drift. Ignores noise (small position deltas, createdAt/updatedAt). Pair with workflow.get to compare deployed vs local. Deterministic.
execution.explainDiagnose a failed or surprising n8n execution. Paste the execution JSON (from the n8n UI 'Show details' or `GET /executions/:id?includeData=true`); returns a per-node summary highlighting nodes that returned 0 items, unresolved `={{ ... }}` expressions, errors with hints, and LLM token usage. Hits the most common debugging pain point: items 'silently disappearing' between nodes. Deterministic, rule-based.
execution.replayBuild a self-contained replay workflow that exercises a single node from a larger workflow. The replay workflow is Manual Trigger -> Replay Seed (Code node with pinned items) -> target node. Optional `inputItems` or an `execution` payload pins what the target sees. Useful for iterating on one stubborn node without re-running the whole pipeline. Returns workflow JSON ready to import or push via workflow.create.
execution.timelineRender an n8n execution as a per-node timeline: start offset, duration, items in/out, error flag. Complements execution.explain — that one surfaces *why*, this surfaces *when*. Output is a markdown table sorted by start time. Deterministic.
workflow.listList workflows from a live n8n instance (requires N8N_API_URL + N8N_API_KEY env vars). Returns id, name, active, nodeCount, updatedAt, tags. Filter by active, tags, name. Use this when the user asks 'what workflows do I have?' or before workflow.get.
workflow.getFetch a single workflow JSON by id from a live n8n instance (requires N8N_API_URL + N8N_API_KEY). Returns the full nodes/connections payload — pair with workflow.lint to audit a deployed workflow.
workflow.createCreate a workflow on a live n8n instance (requires N8N_API_URL + N8N_API_KEY). Strips read-only fields (id, active, createdAt, ...) before posting. Workflows are created inactive — call workflow.activate afterward. Pairs with workflow.generate for end-to-end 'describe -> deploy'.
workflow.activateActivate or deactivate a workflow on a live n8n instance (requires N8N_API_URL + N8N_API_KEY). Pass `active: false` to deactivate. Idempotent — re-activating an already-active workflow is a no-op on n8n's side.
execution.listList recent executions from a live n8n instance (requires N8N_API_URL + N8N_API_KEY). Filter by workflowId, status (success|error|waiting), limit. Pass `includeData: true` to get the full execution body (large) — pair with execution.explain to diagnose a specific failure.