Connect your n8n instance to Claude Code — search 9,820 real templates, validate drafts offline, deploy to self-hosted n8n, and auto-repair failed runs.
For: Automation Leads, RevOps Builders, Integration Consultants, AI-First Operations Teams
Sign in to start chatting with this dataset.
Same data, different audiences.
Use the same data from Claude, Cursor or ChatGPT. AI agents can call every tool the app uses.
Open Chat & MCPMCP for self-hosted n8n: deploy, run, and auto-repair workflows on your instance, backed by a materialized node and template knowledge index.
Searching templates and checking drafts works without your instance. Deploying, running, and inspecting live workflows requires a connected n8n server with an access key from n8n Settings.
MCPBundles maintains a searchable index built from the public n8n template library and node catalog, refreshed on a schedule so agents query local evidence instead of hammering the public site on every turn.
Yes. Drafts can be checked offline for missing node names, bad wiring, and common expression mistakes. After deploy, live checks and run inspection help repair failures without rebuilding from scratch.
Ops and automation leads who already use n8n but want agents to search real examples, assemble drafts, and push to a self-hosted instance — not developers wiring custom integrations from memory.
Enable n8n Automation MCP on MCPBundles, add your instance URL and access key from n8n Settings, then ask in chat. Template search works immediately; deploy steps use the saved connection.
Same dataset, exposed as a Model Context Protocol server. Drop the URL into Claude Desktop, Cursor, VS Code, Cline, Windsurf, or any other MCP client.
https://mcp.mcpbundles.com/bundle/n8n-automation-mcpThe link prefills the Add custom connector dialog — you still review the values and click Add, then Connect to complete OAuth.
n8n Knowledge Index and paste the MCP URL into Remote MCP server URL.Custom connectors at claude.ai require a paid Claude plan (Pro, Max, Team, or Enterprise).