The Code Review Agent is a built-in capability in Agent Mode for automated code review. It can perform deep code reviews across your specified scope — entire projects, specific files, Git diffs, or Pull Requests — providing improvement suggestions across multiple dimensions including code quality, potential defects, security risks, performance issues, readability, and best practices.
Use the /code-review command in a conversation or describe your review requirements in natural language, and the agent will automatically analyze the target code and generate a structured review report.
Core capabilities
- Multi-dimensional review: Detects logic defects, security vulnerabilities (SQL injection, XSS, etc.), performance bottlenecks, code style issues
- Flexible review scope: Supports full project scanning, specific files, Git diff changes, Pull Request reviews
- Structured reports: Categorized by severity (error, warning, suggestion), with specific code locations, issue descriptions, and fix recommendations
How to use
The Code Review Agent is built into Agent Mode — no separate setup required. Open the chat panel, switch to Agent Mode, and use the /code-review command with your desired review scope.
For large changes, review in batches by module for more granular feedback. If changes involve specific business logic, briefly explain the context in your request for more relevant suggestions.
