Pydantic AI Docs

Provides programmatic access to Pydantic AI documentation, including cloning/updating docs, retrieving specific documents, listing topics, and viewing changelogs.

python

3

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "omniwaifu-pydantic-ai-docs-server": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/your/pydantic-ai-docs-server",
        "run",
        "-m",
        "pydantic_ai_docs_server"
      ]
    }
  }
}

You can access and control the Pydantic AI documentation through an MCP server. This setup lets you clone or update the docs, fetch specific documents, list topics, and view changelogs programmatically, all via standardized MCP tools. This makes it easy to integrate docs data into your apps or workflows.

How to use

Use an MCP client to discover the available tools and then call them to interact with the documentation data. The server operates over a standard input/output channel, accepting JSON requests and returning JSON responses. Key capabilities include updating the documentation repository, retrieving a document by path, listing docs topics, and accessing changelogs.

How to install

Prerequisites: Python 3.12 or newer, and a tool for running the MCP server (uv) or a Python environment you can activate.

Clone this project and enter the directory.

Create and activate a Python virtual environment.

python -m venv .venv
source .venv/bin/activate  # On Windows use: .venv\Scripts\activate

Install dependencies in editable mode so you can run and develop locally.

uv pip install -e .
# Or, if you don't have uv:
# pip install -e .

Additional notes

Run the server after installation. You can start it by running the module directly or by using the installed script. The server is designed to communicate over stdio using newline-delimited JSON messages, making it straightforward to integrate with MCP clients.

Available tools

update_documentation

Clones the documentation repository if not present or pulls the latest updates. If force_clone is true, it deletes the existing repository and clones it anew.

get_document_by_path

Retrieves a specific documentation file by its path relative to the docs directory, for example usage/models.md.

list_topics

Lists files and directories within the docs directory. If a path is provided, lists contents of that subdirectory.

list_available_changelogs

Lists all changelog files found under the docs history folder in the repository.

get_changelog_content

Retrieves the content of a specific changelog file, such as history/0.2.0.md.