gemini-kit
π 19 AI Agents + 44 Commands for Gemini CLI - Code 10x faster with auto planning, testing, review & security
claude mcp add --transport stdio nth5693-gemini-kit npx -y gemini-kit
How to use
Gemini-Kit is an extension for the Gemini CLI that transforms your terminal into an AI-powered engineering team. It provides 27 specialized AI agents, 45 slash commands, and 33 workflows to help you design, plan, implement, test, and review software more quickly. When you run the MCP server, youβll be able to invoke agents (e.g., security auditors, frontend specialists, backend specialists, DevOps engineers, and more) and orchestrate multi-agent workflows to research topics, generate plans, implement features, and perform reviews. Use commands like /plan, /review, /cook, and /code to drive end-to-end development with the integrated agent ecosystem. The extension emphasizes compound workflows that iterate to build a knowledge base, with learning and security hooks built in to streamline collaboration across tasks.
How to install
Prerequisites:
- Node.js 18 or newer
- Git
- Gemini CLI installed and configured
Installation options: Option A β Quick MCP server setup (recommended with Gemini CLI):
- Ensure you have Node.js and Git installed and that Gemini CLI is accessible in your PATH.
- Install the Gemini-Kit MCP server via npx (no global install required): npm pkgManager = npm; npm install -g npm@latest npx -y gemini-kit
Option B β Install as a local extension (works with your project):
- From a project directory, install the package locally: npm install gemini-kit --save-dev
- Use the MCP server runner to start it via your existing MCP tooling or by invoking npx as above.
Notes:
- The Quick Start in the project docs demonstrates how to clone and build Gemini-Kit as an extension. For MCP usage, you generally invoke the npm/package via npx with the provided command structure.
- If you prefer a global install, you can run: npm install -g gemini-kit
- After installation, verify the package by listing installed commands and ensuring the Gemini CLI recognizes the extension.
Additional notes
Tips and common issues:
- Ensure Node.js version compatibility (β₯18).
- If you encounter permission errors during npm install, try using a node version manager (e.g., nvm) and avoid sudo.
- The MCP configuration uses the npx invocation with the package name; if you publish or rename the package, update the mcp_config accordingly.
- Gemini-Kit relies on the Gemini CLI ecosystem; ensure Gemini CLI is installed and up-to-date for best compatibility.
- When running in CI, pin the Node version and Git to reproducible states and consider caching node_modules between builds.
- If you see agent or command parsing issues, consult the MCP server logs for which agent pull/request failed and verify the corresponding command syntax in the Gemini-Kit docs.
Related MCP Servers
sandboxed.sh
Self-hosted orchestrator for AI autonomous agents. Run Claude Code & Open Code in isolated linux workspaces. Manage your skills, configs and encrypted secrets with a git repo.
automagik-genie
π§ Automagik Genie β bootstrap, update, and roll back AI agent workspaces with a single CLI + MCP toolkit.
sub-agents
Define task-specific AI sub-agents in Markdown for any MCP-compatible tool.
mcpresso
TypeScript framework to build robust, agent-ready MCP servers around your APIs.
kratos
ποΈ Memory System for AI Coding Tools - Never explain your codebase again. MCP server with perfect project isolation, 95.8% context accuracy, and the Four Pillars Framework.
codingbuddy
Codingbuddy orchestrates 29 specialized AI agents to deliver code quality comparable to a team of human experts through a PLAN β ACT β EVAL workflow.