LabscoMCP early access ↗
blesscatMCP SERVER

Gemini CLI MCP by blesscat

Ask Gemini a question from inside another assistant, through the Gemini CLI already set up on your machine.

Model Routing, Multi-Model Consultation & Cost ControlCommunity

LABSCO SUMMARY

A second opinion from Gemini, on the credentials your CLI already holds.

Because it shells out to the Gemini CLI, there is no separate API key to manage: whatever your CLI is authorised to do, the server can do. Three tools, all of them straight passthrough.

What it is

A server that wraps the Gemini CLI: chat, generation with temperature control, and the list of available models.

What you get

  • A message sent to Gemini and its reply returned, with the model selectable and gemini-pro as the default (gemini_chat)
  • Content generated from a prompt, with a model and a temperature you choose (gemini_generate)
  • The Gemini models your CLI credentials can reach (gemini_list_models)

Requirements

Node.js 18 or newer, and the Gemini CLI installed, credentialed and on your PATH. If the gemini command does not run in your shell, neither does the server.

Setup effort

Build from source — git clone https://github.com/blesscat/gemini-cli-mcp