AI Code Review Prompts

Review this pull request: [paste code diff or files]. Check for: bugs and logic errors, security vulnerabilities (injection, auth, data exposure), performance issues (N+1 queries, memory leaks, unnecessary computation), code style and naming inconsistencies, missing error handling, and test coverage gaps. Categorize findings as: critical, important, nitpick. Suggest fixes for critical and important items.

Audit this code for security vulnerabilities: [paste code]. Check for: SQL/NoSQL injection, XSS possibilities, authentication/authorization bypasses, sensitive data exposure (logs, errors, responses), insecure dependencies, CSRF vulnerabilities, and misconfigured CORS. For each finding: describe the attack vector, rate severity (critical/high/medium/low), and provide the fix.

Review this code specifically for performance: [paste code]. Context: this runs [frequency - e.g., "per API request" or "once daily"]. Data scale: [approximate volumes]. Identify: algorithmic inefficiencies (O(n) that could be O(1)), unnecessary database queries, memory allocation issues, missing caching opportunities, and expensive operations that could be deferred. Quantify impact where possible.

Review these tests for completeness: [paste tests + code being tested]. Identify: untested code paths, missing edge cases, tests that would pass even if the code was broken (tautological tests), missing error condition tests, and tests that are too coupled to implementation details. Suggest the 5 most valuable tests to add, ordered by risk reduction.

Review this codebase for architectural consistency: [paste code or describe structure]. Patterns in use: [MVC, repository, etc.]. Check: are patterns applied consistently, does the abstraction level make sense, are there circular dependencies, is the code organized for testability, and where does complexity concentrate? Suggest specific refactoring for the worst offenders.