-
Notifications
You must be signed in to change notification settings - Fork 4
Code Review Workflow
Alessio Rocchi edited this page Jan 27, 2026
·
1 revision
Complete pattern for conducting code reviews with multiple agents.
Use Researcher, Reviewer, and optionally Adversarial agents to perform comprehensive code reviews.
sequenceDiagram
participant User
participant Researcher
participant Reviewer
participant Adversarial
User->>Researcher: Analyze changes
Researcher-->>User: Context & impact
User->>Reviewer: Review code
Reviewer-->>User: Quality feedback
User->>Adversarial: Security review
Adversarial-->>User: Security issues
const session = await memory.createSession({
type: 'code-review',
pr: 'PR-123',
repository: 'my-app'
});Spawn a researcher agent to analyze the changes in PR-123
Spawn a reviewer agent to review the code for quality and best practices
For security-critical changes:
Spawn an adversarial agent to find security vulnerabilities
Combine all findings and create review summary
Complete code review for authentication PR:
1. Session: code-review-auth-pr
2. Researcher: Analyzes auth changes
3. Reviewer: Checks code quality
4. Adversarial: Tests for vulnerabilities
5. Result: Comprehensive review report
Related:
Getting Started
Core Concepts
Agent Guides
- Overview
- Coder
- Researcher
- Tester
- Reviewer
- Adversarial
- Architect
- Coordinator
- Analyst
- DevOps
- Documentation
- Security Auditor
MCP Tools
- Overview
- Agent Tools
- Memory Tools
- Task Tools
- Session Tools
- System Tools
- GitHub Tools
- Review Loop Tools
- Identity Tools
Recipes
- Index
- Code Review
- Doc Sync
- Multi-Agent
- Adversarial Testing
- Full-Stack Feature
- Memory Patterns
- GitHub Integration
Advanced
- Plugin Development
- Custom Agent Types
- Workflow Engine
- Vector Search Setup
- Web Dashboard
- Programmatic API
- Resource Monitoring
Reference