code-whisperer

MCP server that gives AI-powered code analysis to any MCP client. Uses Groq for fast LLM inference.

Tools

• review_code — check for bugs, security issues, style problems, performance concerns
• explain_code — explain code in plain English
• optimize_code — suggest speed and readability improvements
• generate_tests — generate pytest unit tests
• ask_groq — ask Groq a programming question

Usage

python -m venv venv
source venv/bin/activate
pip install -r requirements.txt
cp .env.example .env   # add your GROQ_API_KEY
code-whisperer

Opencode config (opencode.json):

{
  "mcp": {
    "code-whisperer": {
      "type": "local",
      "command": ["/bin/bash", "-c", "cd /path/to/code-whisperer && exec /path/to/code-whisperer/venv/bin/code-whisperer"],
      "enabled": true
    }
  }
}

Claude Desktop config (claude_desktop_config.json):

{
  "mcpServers": {
    "code-whisperer": {
      "command": "/bin/bash",
      "args": ["-c", "cd /path/to/code-whisperer && exec /path/to/code-whisperer/venv/bin/code-whisperer"]
    }
  }
}

Config

Env var	Required	Default
GROQ_API_KEY	yes	—
GROQ_MODEL	no	llama-3.3-70b-versatile