Package Exports

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (@mugzie/ai-council) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

AI Council

Multi-agent AI code review and decision engine. Get consensus from multiple AI perspectives on your code changes.

Installation

npm install -g @mugzie/ai-council

License

A license key is required. Purchase at ai-council.lemonsqueezy.com.

export AI_COUNCIL_LICENSE_KEY="your-license-key"

Usage

# Review current git diff
ai-council review --diff --branch=main --pretty

# Specialized reviews
ai-council security --diff --pretty   # Security-focused review
ai-council perf --diff --pretty       # Performance review
ai-council arch --diff --pretty       # Architecture review
ai-council sanity --diff --pretty     # General sanity check

# CI mode (exits with code 1 on REJECT)
ai-council review --diff --ci

MCP Server (Claude / Cursor)

ai-council mcp

Claude Code

claude mcp add --transport stdio ai-council -- ai-council mcp

Cursor

Add to .mcp.json:

{
  "mcpServers": {
    "ai-council": {
      "command": "ai-council",
      "args": ["mcp"],
      "env": {
        "AI_COUNCIL_LICENSE_KEY": "your-license-key",
        "OPENAI_API_KEY": "your-openai-key",
        "GEMINI_API_KEY": "your-gemini-key"
      }
    }
  }
}

Environment Variables

Variable	Required	Description
`AI_COUNCIL_LICENSE_KEY`	Yes	Your license key
`OPENAI_API_KEY`	No	OpenAI API key
`GEMINI_API_KEY`	No	Google Gemini API key

How It Works

AI Council assembles a panel of AI agents with different perspectives:

Senior Developer - Code quality and best practices
Security Engineer - Security vulnerabilities
Performance Engineer - Performance implications
Architect - Design patterns and structure
Gemini - Alternative perspective (Google's Gemini)

Each agent votes (APPROVE / REVISE / REJECT) with confidence scores. A judge agent synthesizes the votes into a final recommendation.

@mugzie/ai-council