Skip to content
MCP server

Gemini Pro MCP

Use your Google AI Pro plan from Claude, Codex and Gemini CLI through MCP tools

Runs locallyStatus and history
Install
Pricing
Free
Platforms
CLI, macOS, Windows
Runs
Locally
Version
0.1.5
Listing updated
today

Similar MCP servers

About Gemini Pro MCP

Gemini Pro MCP is a free MCP server for ai models and llm tooling. It runs locally on your machine and is started by your AI client.

Rather than a paid Gemini API key, the Gemini tools here run through the Antigravity CLI signed in to your Google AI Pro or Ultra account, so they draw on the plan's own limits. A single setup command detects Claude Code, Codex, the Antigravity CLI and the Gemini CLI and adds the server to each, and a doctor command reports what works and how to fix what does not.

The tool set is grouped by what it runs on. Gemini and Deep Research calls use the subscription, Jules tools use an optional Jules API key, and image and video tools use the Gemini API, which is billed per use. Experimental browser tools drive your own signed-in Chrome to reach Google Flow, Gemini Notebooks and AI Mode; they are off until you enable them, and the README says the server never sees your Google password.

Defaults lean cautious: Gemini reads in a sandbox and cannot change files unless you call the code tool, and paid calls ask first unless you pre-approve spending.

Install

Run in a terminal

claude mcp add gemini-pro-mcp --env GEMINI_API_KEY='<GEMINI_API_KEY>' --env JULES_API_KEY='<JULES_API_KEY>' -- uvx gemini-pro-mcp

Add to claude_desktop_config.json

{
  "mcpServers": {
    "gemini-pro-mcp": {
      "command": "uvx",
      "args": [
        "gemini-pro-mcp"
      ],
      "env": {
        "GEMINI_API_KEY": "<GEMINI_API_KEY>",
        "JULES_API_KEY": "<JULES_API_KEY>"
      }
    }
  }
}

Add to ~/.cursor/mcp.json

{
  "mcpServers": {
    "gemini-pro-mcp": {
      "command": "uvx",
      "args": [
        "gemini-pro-mcp"
      ],
      "env": {
        "GEMINI_API_KEY": "<GEMINI_API_KEY>",
        "JULES_API_KEY": "<JULES_API_KEY>"
      }
    }
  }
}

Add to .vscode/mcp.json

{
  "servers": {
    "gemini-pro-mcp": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "gemini-pro-mcp"
      ],
      "env": {
        "GEMINI_API_KEY": "<GEMINI_API_KEY>",
        "JULES_API_KEY": "<JULES_API_KEY>"
      }
    }
  }
}

Add to ~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "gemini-pro-mcp": {
      "command": "uvx",
      "args": [
        "gemini-pro-mcp"
      ],
      "env": {
        "GEMINI_API_KEY": "<GEMINI_API_KEY>",
        "JULES_API_KEY": "<JULES_API_KEY>"
      }
    }
  }
}

Add to your client's MCP settings (most use this shape)

{
  "mcpServers": {
    "gemini-pro-mcp": {
      "command": "uvx",
      "args": [
        "gemini-pro-mcp"
      ],
      "env": {
        "GEMINI_API_KEY": "<GEMINI_API_KEY>",
        "JULES_API_KEY": "<JULES_API_KEY>"
      }
    }
  }
}

Using several servers? Build one config file with the MCP config generator.

Configuration

  • GEMINI_API_KEYGemini API key for the video and image tools (optional; these tools are billed)
  • JULES_API_KEYJules API key from jules.google.com/settings (optional, for the jules_* tools)
  • GEMINI_PRO_ENABLE_FLOWSet to 1 to enable the Google Flow tools
  • GEMINI_PRO_ENABLE_BROWSERSet to 1 to enable all browser tools (Gemini app, Notebooks, AI Mode)

Tools this server provides (31)

Taken from the project documentation; the running server may differ.

gemini_ask
No description
gemini_research
No description
gemini_code
No description
gemini_agents
No description
gemini_models
No description
jules_start
No description
jules_status
No description
jules_approve_plan
No description
jules_message
No description
jules_sources
No description
jules_sessions
No description
video_generate
No description
video_edit
No description
video_check
No description
image_generate
No description
video_stitch
No description
gemini_app
No description
gemini_app_result
No description
google_ai_mode
No description
google_page
No description
google_act
No description
google_pro_benefits
No description
flow_make_video
No description
flow_job
No description
flow_generate
No description
flow_approve
No description
flow_reject
No description
flow_collect
No description
flow_status
No description
flow_tools
No description
flow_open_tool
No description

What it does

Gemini on your plan
Ask, research and code tools run through the Antigravity CLI instead of a pay-per-token key.
Jules coding agent
Start sessions, approve plans and message Jules with a Jules API key.
Media generation
Video and image tools use the Gemini API and are billed per use.
Flow and Notebooks
Optional browser tools reach Google Flow and Gemini Notebooks.
One-command setup
A setup command configures each detected client and doctor checks the result.

Good for

  • Get a second opinion from Gemini on a source file
  • Run sourced research from inside a coding agent
  • Generate a short video with Flow credits
  • Turn PDFs into a Gemini Notebook Audio Overview
  • Delegate a repository task to Jules

Pricing

Free

Needs a Google AI Pro or Ultra plan; image and video tools bill through the Gemini API

Gemini Pro MCP FAQ

Do I need a Gemini API key?

Only for the image and video tools. The Gemini tools use your AI Pro plan through the Antigravity CLI.

How do I install it?

Run uvx gemini-pro-mcp setup.

Are the browser tools on by default?

No, they are off until you enable the browser option.

Does it spend money without asking?

Paid calls ask first unless you use the auto-approve option.

How many tools are there?

Up to 36, according to the README.

Is Gemini Pro MCP free?

Yes. Gemini Pro MCP is free to use.

How do I add Gemini Pro MCP to Claude Code?

Run this command in your terminal: claude mcp add gemini-pro-mcp --env GEMINI_API_KEY='<GEMINI_API_KEY>' --env JULES_API_KEY='<JULES_API_KEY>' -- uvx gemini-pro-mcp

What are the best alternatives to Gemini Pro MCP?

The closest alternatives to Gemini Pro MCP are headroom, omniget, next-ai-draw-io, mcp-for-blender and Claude Code Ultimate Guide.

Where this comes from

  • Description: written from the project's README.
  • Install steps: from the official MCP Registry entry.

How AIPoweredBot checks servers