AI Prompts for Code Review

Review the following {{language}} code for security vulnerabilities. Focus on the OWASP Top 10 categories relevant to this stack, including:

- SQL injection and NoSQL injection vectors
- Cross-site scripting (XSS) — stored, reflected, and DOM-based
- Insecure deserialization
- Broken authentication or session management
- Sensitive data exposure (hardcoded secrets, logging PII)

For each vulnerability found, provide:
1. The exact line(s) affected
2. The severity (Critical / High / Medium / Low)
3. A concrete fix with code

Context:
- Framework: {{framework}}
- This code handles: {{code_purpose}}

```
{{code}}
```

Analyze the following {{language}} code for performance issues. Consider the production context: this code runs in {{deployment_context}} handling approximately {{expected_load}}.

Review for:
- Time complexity issues (nested loops, unnecessary iterations)
- N+1 query patterns or redundant database calls
- Memory allocation and potential leaks
- Blocking operations that should be async
- Missing caching opportunities
- Inefficient data structures for the access pattern

ORM/Database: {{orm_and_database}}

For each issue, estimate the performance impact (orders of magnitude) and provide an optimized alternative.

```
{{code}}
```

Review this {{language}} module for architectural quality. The codebase follows {{architecture_pattern}} and this module is responsible for {{module_responsibility}}.

Evaluate:
- Single Responsibility: Does each class/function do one thing well?
- Dependency direction: Are dependencies pointing inward (toward domain logic)?
- Abstraction levels: Is the code mixing high-level orchestration with low-level details?
- Interface boundaries: Are the public APIs clean and minimal?
- Extensibility: Can new behavior be added without modifying existing code?
- Testability: Can this code be unit tested without complex mocking?

Flag any violations of {{team_conventions}} if present.

Provide a summary rating (Strong / Acceptable / Needs Refactoring) with prioritized recommendations.

```
{{code}}
```

Act as a senior engineer reviewing this pull request. The PR is: {{pr_description}}

Review the diff below against this checklist:

**Correctness**
- Does the logic match the stated intent?
- Are edge cases handled (null, empty, boundary values)?
- Are error paths handled gracefully?

**Readability**
- Are variable and function names descriptive?
- Is the code self-documenting or does it need comments?
- Is there unnecessary complexity?

**Testing**
- Are the changes covered by tests?
- What test cases are missing?

**Side Effects**
- Could this break existing functionality?
- Are there backward compatibility concerns?
- Does this affect any shared state or global config?

Output a structured review with APPROVE, REQUEST_CHANGES, or COMMENT for each section and specific inline feedback referencing line numbers.

```diff
{{diff}}
```

Analyze the following {{language}} code for technical debt. This code is {{code_age}} and is maintained by a team of {{team_size}}.

Categorize any tech debt found into:
1. **Critical** — Will cause production incidents if not addressed
2. **Strategic** — Slows down feature development significantly
3. **Cosmetic** — Code smell but low practical impact

For each item found, provide:
- What the debt is and where it lives
- Why it accumulated (likely cause)
- Effort estimate to fix (hours: 1-2h, 4-8h, multi-day)
- Suggested approach (quick patch vs. proper refactor)
- Risk of the fix introducing regressions

Conclude with a prioritized action plan that respects the team's bandwidth: what to fix this sprint, what to schedule, and what to accept.

```
{{code}}
```

Review this {{language}} code against our team's style guide. Enforce the following conventions strictly:

{{style_rules}}

For each violation:
- Quote the offending line
- State which rule it violates
- Provide the corrected version

Do NOT flag issues outside of the rules above — focus only on style consistency, not logic or architecture. If the code fully complies, say so explicitly.

Group violations by rule for easy scanning. At the end, provide a single copy-pasteable corrected version of the full code.

```
{{code}}
```