Senior Code Reviewer

What it does

Senior Code Reviewer is a Cursor agent playbook named code-reviewer. AgenticKit installs it so a session can adopt this role instead of acting as a generic coding assistant.

The code-reviewer agent provides instant, high-signal feedback on pull requests. It checks for memory leaks, unhandled promise rejections, accessibility issues, missing tests, and architectural consistency with CURSOR.md.

Conducts strict PR code reviews for readability, performance, and best practices. Senior code reviewer agent for Cursor that inspects git diffs, catches edge cases, prevents performance regressions, and suggests clean refactors.

Why it exists

A pile of agent files is not a team. code-reviewer exists so one job, senior code reviewer, has a written brief, required skills, and a stop condition.

It ships in the Engineering Kit. Skills it must read: Project Conventions & CURSOR.md Standards; Strict TypeScript Standards & Type Safety; Vitest Unit & Integration Testing. Commands that adopt this role: /audit: Codebase Quality & Security Auditor; /ship: End-to-End Feature Pipeline Shipper; /fix: Automated Bug & Exception Debugger.

When to use it

• Pre-merge code reviews on complex pull requests
• Verifying test coverage and edge case handling
• Maintaining high architectural standards across solo projects

When not to use it

• Do not ask code-reviewer to do a different role's job. If you need a launch post, switch agents.
• Do not keep the same chat after this agent has finished its artifact. Start a reviewer in a new thread.
• Do not invoke every agent in the kit for a small change.

Example workflow

1. Install the Engineering Kit so .cursor/agents/code-reviewer is on disk.
2. Run /audit: Codebase Quality & Security Auditor, or start a chat and tell Cursor to adopt the code-reviewer role.
3. The agent should read: Project Conventions & CURSOR.md Standards; Strict TypeScript Standards & Type Safety; Vitest Unit & Integration Testing.
4. It produces the artifact for this role only, then stops.
5. A different agent or you review. Same-chat self-review is not a review.

Example usage

Example: "You are code-reviewer. Pre-merge code reviews on complex pull requests. Read project-conventions before you edit."

Or let the pipeline invoke it: /audit: Codebase Quality & Security Auditor.

Capabilities you should actually see: Line-by-line pull request diff reviews and risk scoring

Example output

• code-reviewer should leave files or a written verdict, not a vibe check. Line-by-line pull request diff reviews and risk scoring Detection of anti-patterns, race conditions, and dead code
• Engineering agents should touch the slice they were given (schema, route, test, or review note) and nothing else.

Best practices

• Line-by-line pull request diff reviews and risk scoring
• Detection of anti-patterns, race conditions, and dead code
• Enforcement of team coding style and TypeScript conventions
• Constructive, actionable refactoring suggestions
• One role per chat unless a command is explicitly orchestrating a sequence.

Common mistakes

• Using code-reviewer as a synonym for 'the Cursor agent'. It is a brief, not the product.
• Skipping the required skills and hoping the role name is enough.
• Letting the writer approve its own PR.
• Invoking this agent and three unrelated ones in the same prompt.