Graphor Remote MCP Server

The Graphor Remote MCP Server lets you connect AI assistants and agentic frameworks to the Graphor API using the Model Context Protocol (MCP). It runs on Cloudflare Workers and uses OAuth for authentication.

Server URL:

https://mcp.graphor.workers.dev/sse

Looking for the local MCP server instead? See the Local MCP Server section below or the graphor-mcp package.

Local MCP Server

For local MCP clients (e.g. Cursor, VS Code), you can use the graphor-mcp package which authenticates via API key instead of OAuth.

Install directly using the buttons below:

 

Or manually add it to your MCP client's configuration:

{
  "mcpServers": {
    "graphor_api": {
      "command": "npx",
      "args": ["-y", "graphor-mcp@latest"],
      "env": {
        "GRAPHOR_API_KEY": "grlm_your_api_key_here"
      }
    }
  }
}

Note: You can get your API key from graphorlm.com.

Remote MCP Server

For web-based AI clients (e.g. Claude.ai) or agentic frameworks (e.g. LangChain, CrewAI) that cannot run local npx processes, use the hosted remote MCP server. Authentication is handled via OAuth — a browser window will open for you to log in.

Claude.ai

1. Go to Settings > Connectors > Add custom connector
2. Fill in the Name (e.g. "Graphor")
3. Set the Remote MCP server URL to:

https://mcp.graphor.workers.dev/sse

1. You will be redirected to log in through the OAuth flow
2. Once authenticated, Graphor tools will be available in your conversations

Claude Desktop

Add the following to your Claude Desktop configuration file (Settings > Developer > Edit Config):

{
  "mcpServers": {
    "graphor_api": {
      "command": "npx",
      "args": ["mcp-remote", "https://mcp.graphor.workers.dev/sse"]
    }
  }
}

When you open Claude Desktop, a browser window will open for you to log in. After authenticating, the Graphor tools will appear in the bottom right of your conversation.

Cursor

Add the following to your Cursor MCP configuration (Settings > Tools & MCP > New MCP Server):

{
  "mcpServers": {
    "graphor_api": {
      "command": "npx",
      "args": ["mcp-remote", "https://mcp.graphor.workers.dev/sse"]
    }
  }
}

Agentic Workflows (LangChain, CrewAI, etc.)

For agentic frameworks that support MCP, connect to the remote server via SSE transport:

LangChain (Python):

from langchain_mcp_adapters.client import MultiServerMCPClient

async with MultiServerMCPClient(
    {
        "graphor": {
            "url": "https://mcp.graphor.workers.dev/sse",
            "transport": "sse",
        }
    }
) as client:
    tools = client.get_tools()
    # Use tools with your LangChain agent

LangChain (TypeScript):

import { MultiServerMCPClient } from "@langchain/mcp-adapters";

const client = new MultiServerMCPClient({
  graphor: {
    url: "https://mcp.graphor.workers.dev/sse",
    transport: "sse",
  }
});

const tools = await client.getTools();
// Use tools with your LangChain agent

Note: The OAuth flow will open a browser window on first connection. For headless environments, you may need to complete the OAuth flow beforehand or use mcp-remote as a local proxy.

Any MCP-compatible Client

For any client that supports remote MCP servers via SSE, use the URL:

https://mcp.graphor.workers.dev/sse

For clients that only support stdio transport, use mcp-remote as a bridge:

{
  "mcpServers": {
    "graphor_api": {
      "command": "npx",
      "args": ["mcp-remote", "https://mcp.graphor.workers.dev/sse"]
    }
  }
}

Troubleshooting

If you run into issues connecting, try the following:

1. Restart your client (Claude Desktop, Cursor, etc.)

1. Test the connection directly on the command line:

npx mcp-remote https://mcp.graphor.workers.dev/sse

1. Clear the OAuth cache if authentication seems stuck:

rm -rf ~/.mcp-auth

Self-hosting

You can deploy your own instance of this server to Cloudflare Workers.

One-click deploy


Manual deploy

1. Clone this repository and install dependencies:

npm install

1. Create the KV namespace:

npx wrangler@latest kv namespace create remote-mcp-server-oauth-kv

1. Add the KV namespace ID to wrangler.jsonc

1. Deploy:

npm run deploy

Local development

npm install
npm run dev

The server will be available at http://localhost:8787/.

To test with the MCP Inspector:

1. Run npx @modelcontextprotocol/inspector
2. Switch Transport Type to SSE and enter http://localhost:8787/sse
3. Click "Connect" and log in through the OAuth screen