Gemini Docs

Provides search and fetch capabilities for Gemini API documentation via an MCP server.

javascript

0

GitHub Stars

javascript

Language

5 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": {
    "leomaiajr-gemini-docs-skills": {
      "command": "node",
      "args": [
        "/path/to/gemini-docs-mcp/dist/index.js"
      ]
    }
  }
}

You run an MCP server that lets you search and fetch Gemini API documentation. It exposes practical tools to quickly locate relevant pages and extract structured content for easy reuse in your workflows.

How to use

You connect to the Gemini Docs MCP Server through an MCP client. Use the available tools to search the Gemini API documentation index and to fetch and parse specific documentation pages with structured content. The server runs as a local process that you start and then query using your MCP client. When you initiate a search, you specify your query and optional limits, and you’ll receive a list of relevant pages. When you fetch a page, you provide the path to the documentation (or leave it empty for the overview) and choose whether you want the response in markdown or json.

How to install

Prerequisites: you need Node.js and npm installed on your machine.

Install dependencies and build the MCP server packages.

Run the server in stdio mode and keep it running to serve requests from your MCP client.

npm install
npm run build

Additional sections

Configuration for your MCP client is provided for Claude Desktop. You can add the server as a local stdio process by specifying the node command and the path to the built MCP index.

{
  "mcpServers": {
    "gemini-docs": {
      "command": "node",
      "args": ["/path/to/gemini-docs-mcp/dist/index.js"]
    }
  }
}

Available tools

search_gemini_docs

Searches through the Gemini API documentation index and returns pages relevant to your query. You can specify the maximum number of results and the output format (markdown or json).

fetch_gemini_doc

Fetches and parses a specific Gemini API documentation page by its path. You can request the content in markdown or json format and use common paths like the main overview, quickstart, function-calling, embeddings, and more.