Gemini Docs MCP Server

A Model Context Protocol (MCP) server that provides AI assistants with access to the Gemini API documentation.

Features

📖 Access Documentation: Read the full content of Gemini API docs directly within your AI chat interface.
🔍 Search Capabilities: Use the search_docs tool to find specific topics, guides, and API references.
⚡ Fast & Local: Runs locally on your machine, serving files from the gemini-api-docs directory.
🛠️ Broad Compatibility: Works with Claude Desktop, Cursor, VS Code (Google Antigravity, GitHub Copilot), and any MCP-compliant client.

Directory Structure

gemini-api-docs/: Contains the Markdown documentation files.
gemini-docs-mcp/: The MCP server implementation (Node.js/TypeScript).

Prerequisites

Node.js: v16 or higher.
NPM: Installed with Node.js.

Installation

Clone the repository (if you haven't already):

git clone <repository-url>
cd mcp-gemini-docs

Install Dependencies & Build: Navigate to the server directory and build the project:
```
cd gemini-docs-mcp
npm install
npm run build
```

Configuration

Add the server to your MCP client configuration. Replace /absolute/path/to/ with the actual full path to this repository on your machine.

1. Claude Desktop & Claude Code

Edit %APPDATA%\Claude\claude_desktop_config.json (Windows) or ~/Library/Application Support/Claude/claude_desktop_config.json (macOS).

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

2. Google Antigravity (VS Code)

Open File > Preferences > Settings.
Search for "Antigravity MCP".
Edit mcp_config.json (or use Command Palette: "Antigravity: Open MCP Config").
Add:

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

3. Cursor

Go to Cursor Settings > Features > MCP.
Click "Add New MCP Server".
Name: gemini-docs
Type: command
Command: node /absolute/path/to/mcp-gemini-docs/gemini-docs-mcp/build/index.js

4. GitHub Copilot (VS Code)

Note: Requires active Copilot subscription and MCP preview access.

Configure via .vscode/settings.json or your user settings:

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

5. CLI Tools (Gemini CLI, OpenAI Codex CLI)

Pass the server startup command as an argument.

# Example
gemini-cli --mcp-server "node /path/to/mcp-gemini-docs/gemini-docs-mcp/build/index.js"

Usage

Once configured, you can ask your AI assistant questions like:

"Search the Gemini docs for 'structured output'"
"Read the documentation for 'models'"
"How do I use context caching with the Gemini API?"

License

ISC

MCP Servers