mcp

sphinx-mcp is a Python-based MCP server that provides a documentation-oriented interface for MCP tools, prompts, resources, and resource templates used within the Sphinx ecosystem. It serves as a lightweight server component to expose MCP-compliant tooling documentation, enabling you to publish and browse MCP-related capabilities alongside your documentation workflow. Once running, you can leverage the server to explore and reference MCP tooling definitions, prompts, and resources that are relevant to your Sphinx documentation projects. The extension is intended to be used by developers documenting MCP-driven tooling and by teams looking to standardize how MCP assets are presented within documentation.

To use the server, install the Python package, start the server using the provided module entry point, and then connect to the server from your MCP-enabled tooling or documentation pipeline. The server exposes MCP metadata and tooling manifests that you can integrate into your documentation builds or tooling scripts, enabling a consistent, machine-readable view of MCP capabilities within your project’s docs.

Prerequisites:

• Python 3.8+ (or a supported Python runtime)
• pip (or your preferred Python package manager)
• Optional: a virtual environment tool (e.g., venv, virtualenv)

Installation steps:

1. Create and activate a virtual environment (recommended)

   python3 -m venv venv source venv/bin/activate # on Windows use: venv\Scripts\activate

2. Install the sphinx-mcp package from PyPI

   pip install sphinx-mcp

3. Run the MCP server (module entry point)

   python -m sphinx_mcp

4. Verify the server is running by checking the console output or hitting the server URL in a browser or compatible client.

Optional: If you maintain multiple MCP-enabled docs, you can repeat the steps in isolated environments to keep dependencies clean.

Notes and tips:

• This MCP server is in an early stage with limited testing, so expect evolving behavior as the project progresses.
• No explicit environment variables are required to run the default server, but you can configure environment-specific behavior by exporting variables if supported by the sphinx_mcp module in future releases.
• If you encounter issues, verify your Python version compatibility and ensure the sphinx_mcp module is correctly installed in the active environment.
• Since this is primarily a documentation-oriented MCP server for Sphinx, you may want to integrate it alongside your existing Sphinx build steps to ensure MCP metadata is available where needed.
• There is no dedicated npm package for this server since it is Python-based; npm-related details are not applicable here.