Multi-Agent AI System

Code Reviews by AI Council

Get consensus from multiple AI perspectives on your code changes. 6 specialized agents debate, vote, and deliver a unified recommendation.

Get Started Documentation
terminal 
$ ai-council review --diff --branch=main --pretty

🔍 Analyzing git diff against main...

👨‍💻 Senior Developer APPROVE (0.92)
   Code follows best practices, well-structured

🔒 Security Engineer APPROVE (0.88)
   No security vulnerabilities detected

⚡ Performance Engineer REVISE (0.75)
   Consider memoizing the expensive computation
   ┌─ perf ──────────────────────────────────┐
   │ const result = useMemo(() => compute(data), [data]);
   └────────────────────────────────────────┘

🏛️ Architect APPROVE (0.85)
   Good separation of concerns

🤖 Gemini APPROVE (0.90)
   Logic is sound, edge cases handled

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
⚖️  Final Decision: APPROVE (confidence: 0.86)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Features

Everything you need for
intelligent code review

Multi-Agent Consensus

6 specialized AI agents with unique perspectives collaborate to review your code, ensuring comprehensive coverage.

Automated Debates

When agents disagree, they engage in structured debates to defend their positions before the final ruling.

Confidence Scoring

Each vote includes confidence levels. Low-confidence agents automatically abstain, ensuring meaningful votes.

Specialized Reviews

Focus on security, performance, architecture, or general sanity with specialized agent configurations.

CI/CD Integration

Use in CI mode with exit codes for automated pipelines. Block merges on low confidence or rejections.

Fix Suggestions

When agents vote REVISE or REJECT, they can provide concrete code fix examples with syntax-highlighted, prettified output.

AST + Embedding Context

Opt-in pre-review pipeline extracts code units via tree-sitter AST and ranks them with CodeBERT embeddings for focused agent context.

Deep Analysis

File-by-file review with structured findings: line-specific issues, severity levels, fix suggestions, and automatic project-rule injection.

Agent Review

Two-pass Claude Agent SDK pipeline: overview scan then deep per-file review with full codebase access via Read, Grep, and Glob.

MCP Server

Native integration with Claude and Cursor via MCP protocol. Review code directly from your favorite AI tools.

Pricing

Purchase your
AI Council license

Checkout is handled securely by LemonSqueezy. Pick a license, pay, and get your key instantly.

License

Instant delivery via LemonSqueezy

Set in checkout per seat
  • All 6 agents included
  • CLI + CI integration
  • License key delivered instantly
  • Manage seats and billing anytime
Buy License

Need to manage an existing subscription? Use My Account to update billing, seats, or retrieve your license key.

The Council

Meet your
AI Review Board

Each agent brings unique expertise and perspective to every code review

Senior Developer

Code Quality Expert

15+ years experience. Focuses on maintainability, best practices, and long-term sustainability.

Generalist

Security Engineer

Security Specialist

Identifies vulnerabilities, threat modeling, authentication issues, and secure coding practices.

Specialist

Performance Engineer

Performance Specialist

Analyzes algorithmic complexity, resource usage, bottlenecks, and optimization opportunities.

Specialist

Software Architect

Architecture Specialist

Evaluates design patterns, scalability, separation of concerns, and system boundaries.

Specialist

Gemini Thinker

Alternative Reasoning

Powered by Google Gemini. Challenges assumptions and provides unique system-level perspectives.

Challenger

Pragmatist

Practical Problem Solver

Balances perfectionism with shipping. Sometimes "good enough" is the right answer.

Generalist
Process

How the
Council Decides

1

Submit Code

Run ai-council with your git diff or code snippet. The council receives your changes for review.

2

Agents Analyze

Each agent reviews from their unique perspective. Specialists check domain relevance before voting.

3

Cast Votes

Agents vote APPROVE, REVISE, or REJECT with confidence scores. Low confidence triggers abstention.

4

Debate if Needed

If Gemini dissents from majority, agents engage in structured debate defending their positions.

5

Final Ruling

The Judge synthesizes all votes and debates into a final recommendation with rationale.

Ready to elevate your code reviews?

Get started with AI Council today. Install via npm and add your license key.

Purchase License Read the Docs 
# Install globally
npm install -g @mugzie/ai-council

# Set your license key
export AI_COUNCIL_LICENSE_KEY="your-key"

# Start reviewing!
ai-council review --diff --pretty