MCP in Lamatic
Model Context Protocol (MCP) is an open standard for connecting AI assistants to external systems, data sources, APIs, tools, and services through one uniform interface.
Lamatic ships three MCP products, each covering a different direction of the protocol:
- Lamatic MCP is a hosted MCP server at
https://mcp.lamatic.ai/mcp. It lets AI assistants build, deploy, run and debug your flows. - MCP Node lets your flow consume tools from any external MCP server.
- Docs MCP is a hosted MCP server that answers questions about Lamatic itself.
The three MCPs
Lamatic MCP: work on your projects from any AI assistant
Lamatic MCP (formerly Graph MCP) connects Claude, Cursor, VS Code or any MCP client to your Lamatic organization. Add https://mcp.lamatic.ai/mcp to your client and sign in with your Lamatic account. Your assistant can then list projects, create and edit flows, deploy them, run them, read logs, and work with D1 and VectorDB.
Use it when you want an AI assistant to build or drive your flows. For example, a developer in Cursor asking for a new flow, deploying it, and running it to check the result.
MCP Node: consume external MCP servers
The MCP Node connects your Lamatic flow to an existing MCP server. Configure the server’s URL, headers, and protocol once, and any AI node downstream can call the tools that server exposes.
Use it when you want your flow to call a third-party MCP server, such as a weather API, a database, or an internal tool catalog.
Docs MCP: query Lamatic docs from any AI assistant
Docs MCP is a hosted MCP server that Lamatic maintains. It indexes lamatic.ai/docs into a RAG pipeline and exposes a query_docs tool. Any MCP client can ask natural-language questions about Lamatic and get answers with references. No setup, no API key.
Use it when you want your AI assistant to answer questions about Lamatic itself (its nodes, agents, integrations and error messages) instead of you searching the docs manually.
When to use which
| Lamatic MCP | MCP Node | Docs MCP | |
|---|---|---|---|
| Direction | External MCP client → your Lamatic projects | Lamatic flow → external MCP server | External MCP client → Lamatic-hosted docs server |
| Role | Server (hosted by Lamatic) | Client (consumer) | Server (hosted by Lamatic) |
| What you provide | A Lamatic account | The URL of an external MCP server | Nothing |
| What you get | Tools to build, deploy, run and debug flows | External tools available inside the flow | A query_docs tool that answers questions about Lamatic |
| Typical caller | Claude, Cursor, VS Code, custom agents | AI nodes inside your flow | Claude, Cursor, anyone reading the docs |
| Auth model | Sign in with your Lamatic account (OAuth) | Headers and API keys you set on the node | None (free to use) |
| Setup location | Add https://mcp.lamatic.ai/mcp to your client | Studio → MCP/Tools → MCP | Add https://docmcp.lamatic.ai/api/mcp to your client config |
The npm packages @lamatic/graph-mcp and @lamatic/dev-mcp, and the org and project API keys they needed, are deprecated. Use Lamatic MCP instead.
Quick chooser
- “I want Claude or Cursor to build, deploy or run my Lamatic flows.” → Lamatic MCP
- “I want my flow to use tools from an external service via MCP.” → MCP Node
- “I want my AI assistant to answer questions about Lamatic itself.” → Docs MCP
- All of the above. → Connect Lamatic MCP and Docs MCP to your assistant, and use MCP Node inside the flows it builds.