Skip to content

MCP Server

MCP provides an easy way to integrate SerenityGPT functionality within MCP clients such as Cursor, VS Code, or your own MCP client.

This page shows how to run SerenityGPT’s MCP server and connect an MCP-capable client. After setup, you’ll have Serenity tools available directly in your MCP client. Currently, the serenity_docs_answer tool is exposed via MCP.

Prerequisites

  • A running SerenityGPT backend.
  • Docker/Docker Compose on the host where you’ll run the MCP server.
  • An MCP client.

Start the MCP server

Add the following lines to your docker-compose.yaml:

  mcp-server:
    <<: *common
    container_name: mcp-server
    command: python manage.py run_mcp_server --host HOST --port PORT --sse-path /mcp/

Parameters

  • --host — Interface to bind on (e.g., 0.0.0.0 to accept remote connections, or 127.0.0.1 for local only).
  • --port — Port for the MCP server (e.g., 8765).
  • --sse-path — Path where the SSE endpoint is served (default shown: /mcp/).

Example minimal overrides

Start (or restart) Compose:

docker compose up -d mcp-server

Point your client to the MCP server

Cursor (example)

In Cursor go to Settings → Tools & Integrations, click New MCP Server, and paste:

{
  "mcpServers": {
    "sse-server-askme": {
      "url": "HOST:PORT/mcp/", 
      "transport": "sse",
      "headers": {
        "Authorization": "Bearer SERENITY_TOKEN_FOR_THE_RELEVANT_TENANT" 
      }
    }
  }
}

Replace:

  • HOST:PORT with your MCP server host/port (e.g., http://localhost:8765 or https://mcp.example.com depending on your setup/client requirements).
  • SERENITY_TOKEN_FOR_THE_RELEVANT_TENANT with your actual Serenity tenant token.

Use the Serenity tools

Once connected, Serenity tools become available in your client. Currently only the serenity_docs_answer tool is available.

You can invoke the tools by prefacing your question with:

ask serenity: <your question>

Example

  • ask serenity: Show me the docs for connecting Confluence and how to restrict access by space.