Connect to the SavageTech MCP server

The SavageTech MCP server exposes this documentation and the API reference as tools. Connect it from Cursor (or another MCP client) so your AI assistant can look up docs and endpoints while you code.

Works with

OpenAI

…and much more.

Icons from Simple Icons (CC0).

MCP endpoint

Use this URL when adding the server in your client. For production, use your deployed server URL instead of http://localhost:4000.

http://localhost:4000/mcp

Connect by provider

Cursor

Open Settings → Features → MCP, click Add new MCP server, and enter the URL above (or choose Streamable HTTP if asked). Alternatively, edit ~/.cursor/mcp.json or .cursor/mcp.json in your project:

{
  "mcpServers": {
    "savagetech-docs": {
      "url": "http://localhost:4000/mcp"
    }
  }
}

Restart Cursor or reload MCP servers after saving.

Anthropic (Claude)

Claude in the browser (claude.ai): Go to Settings → Connectors → Add custom connector, paste the MCP endpoint URL, and complete the connection.

Claude Desktop: Edit your config (~/Library/Application Support/Claude/claude_desktop_config.json on macOS, or %APPDATA%\\Claude\\claude_desktop_config.json on Windows). Add a remote server entry with the endpoint URL if your client supports it, or see MCP remote servers and Anthropic Custom Connectors for details.

OpenAI (ChatGPT)

Enable Developer mode in ChatGPT Settings → Advanced, then open the Connectors tab and add your MCP server using the endpoint URL. See OpenAI MCP docs for the latest steps.

Google (Gemini)

In Gemini CLI, add the server in your MCP settings (e.g. settings.json or the configured config file) with the endpoint URL and Streamable HTTP transport. See Gemini CLI MCP servers for configuration format.

GitHub (Copilot)

If your organization allows custom MCP servers, add the SavageTech MCP endpoint in Copilot’s MCP settings. Admins can configure access under GitHub Copilot → MCP; see Configure MCP server access.

Hugging Face

Use the endpoint with any MCP-compatible client (Cursor, Claude Desktop, VS Code, etc.). Add the URL in that client’s MCP configuration as above. For Hugging Face Chat and Spaces, follow their client’s instructions for adding a remote MCP server.

Tools provided

Once connected, your assistant can call these tools:

Tool	Description
`list_docs`	List all documentation pages with title and path.
`get_doc`	Get full markdown content for a doc by id (e.g. introduction, tutorials/introduction).
`search_docs`	Search docs by query; returns matching pages with title, path, excerpt.
`openapi_find_path`	Find an API path by prefix and optional method.
`openapi_path_info`	Get operation info for a path and method (summary, parameters, responses).
`openapi_list_by_tag`	List API paths by tag (e.g. Players, Missions, Activity).
`openapi_spec_info`	Get OpenAPI spec title, version, and tags overview.

Other MCP clients

Any client that supports the MCP Streamable HTTP transport can connect to http://localhost:4000/mcp. See the Model Context Protocol docs for details.