Claude Code Gemini

Provides access to Claude Code Gemini models via an OpenAI-compatible API for analysis, brainstorming, and reviews.

javascript

GitHub Stars

javascript

Language

6 months ago

First Indexed

3 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": {
    "shunl12324-claude-code-gemini-mcp": {
      "command": "node",
      "args": [
        "/path/to/claude-code-gemini-mcp/dist/index.js"
      ],
      "env": {
        "GEMINI_MODEL": "gemini-3-pro-preview",
        "GEMINI_API_KEY": "your-api-key",
        "GEMINI_BASE_URL": "https://your-api-endpoint/v1"
      }
    }
  }
}

You can run Claude Code Gemini as an MCP Server to access Gemini models through an OpenAI-compatible API. This server lets Claude Code analyze, brainstorm, and review code via dedicated Gemini endpoints, all orchestrated through MCP client tooling.

How to use

You use an MCP client to connect to the Gemini MCP server. The server exposes tools that let you think through problems, brainstorm ideas, review code, and perform general queries. When you invoke a tool, you pass the required parameters and optional context to guide the Gemini model's behavior. The results come back as structured responses you can incorporate into your workflow.

How to install

Prerequisites: ensure you have Node.js installed on your system. You can verify by running node -v and npm -v.

Clone the MCP server repository, install dependencies, and build the project.

Configuration

Environment variables you must provide to run the server are used to authenticate with the Gemini API and to select the model.

{
  "mcpServers": {
    "gemini": {
      "command": "node",
      "args": ["/path/to/claude-code-gemini-mcp/dist/index.js"],
      "env": {
        "GEMINI_API_KEY": "your-api-key",
        "GEMINI_BASE_URL": "https://your-api-endpoint/v1",
        "GEMINI_MODEL": "gemini-3-pro-preview"
      }
    }
  }
}

Security and access

Treat your API key and endpoint URL as secrets. Do not share the GEMINI_BASE_URL and GEMINI_API_KEY in public places. Use per-project or per-team isolation for credentials when possible.

Troubleshooting and notes

If the server fails to start, verify that Node.js is installed, the build output exists at the specified path, and the environment variables are correctly set in the command you use to launch the MCP server.

Examples and tips

You can place the MCP config in your project under .mcp.json or in your user settings file (e.g., ~/.claude/settings.json). The provided snippet shows a typical configuration for a local Node-based MCP server.

Additional notes

The server exposes tools for deep thinking, brainstorming, reviews, and general queries. Each tool is designed to help you tackle complex problems by leveraging Gemini models through a consistent MCP interface.

Available tools

gemini_think

Deep analysis and reasoning for complex problems.

gemini_brainstorm

Generate multiple creative ideas with pros and cons.

gemini_review

Code, architecture, security, or performance review.

gemini_query

General purpose query with full control.

Built by

VeilStrat

AI signals for GTM teams