AgentCore PR Review Assistant

An AI-powered code review assistant built with AWS AgentCore that automatically analyzes GitHub pull requests for security vulnerabilities, code quality issues, and best practices.

Features

• Security Analysis: Detects hardcoded credentials, injection vulnerabilities, authentication issues
• Code Quality Review: Assesses readability, complexity, naming conventions
• Performance Optimization: Identifies bottlenecks and inefficiencies
• Best Practices: Ensures framework/language conventions
• Actionable Feedback: Provides specific improvement recommendations with code examples

Setup

1. Environment Variables

Create a .env file with:

# AWS Configuration
AWS_ACCESS_KEY_ID=your_access_key
AWS_SECRET_ACCESS_KEY=your_secret_key
REGION_ID=us-west-2

# AgentCore Configuration
MEMORY_ID=your_memory_id
MODEL_ID=anthropic.claude-3-5-haiku-20241022-v1:0

# GitHub Integration (for webhook)
GITHUB_TOKEN=your_github_token
AGENTCORE_RUNTIME_ARN=your_agentcore_runtime_arn

2. Deploy AgentCore Agent

# Install dependencies
pip install -r requirements.txt

# Deploy to AWS
agentcore launch --auto-update-on-conflict \
  --env AWS_ACCESS_KEY_ID=your_key \
  --env AWS_SECRET_ACCESS_KEY=your_secret \
  --env REGION_ID=us-west-2 \
  --env MEMORY_ID=your_memory_id \
  --env MODEL_ID=anthropic.claude-3-5-haiku-20241022-v1:0

3. Test Locally

# Test general assistant
python test_invoke.py

# Test PR review functionality
python test_pr_review.py

4. GitHub Integration (Optional)

Deploy github_webhook.py as an AWS Lambda function and configure GitHub webhooks to automatically review PRs.

Usage

Direct Invocation

# General query
agentcore invoke '{"prompt": "Hello"}'

# PR Review
agentcore invoke '{"pr_data": {"title": "Fix auth bug", "author": "dev", "files": [...]}}'

API Response

{
  "review": "## PR Review Summary\n- Overall assessment\n- Key concerns\n...",
  "pr_title": "Fix auth bug",
  "files_reviewed": 3
}

Architecture

• AgentCore Runtime: Hosts the AI agent with code execution capabilities
• Claude 3.5 Haiku: Provides intelligent code analysis
• Memory Integration: Maintains context across reviews
• GitHub Webhook: Automates PR review process

Security

• No hardcoded credentials in source code
• Environment variables for sensitive configuration
• AWS IAM roles for secure access
• GitHub token for API authentication

Contributing

1. Fork the repository
2. Create a feature branch
3. Make your changes
4. Test thoroughly
5. Submit a pull request

License

MIT License