ai-output-validation
Validates, parses, and sanitizes AI-generated outputs before they reach end users or downstream systems. Structured output enforcement, schema validation, and…
Structured code review focusing on correctness, security, and maintainability. Correctness before style. Every reviewer comment must be actionable.
$ npx -y skills add DevelopersGlobal/ai-agent-skills --skill code-review --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/code-reviewContext preview
The summary Claude sees to decide when to auto-load this skill.
Structured code review focusing on correctness, security, and maintainability. Correctness before style. Every reviewer comment must be actionable.
name: code-review description: Structured code review focusing on correctness, security, and maintainability. Correctness before style. Every reviewer comment must be actionable. category: review applies-to: [claude, gemini, cursor, copilot, any] version: 1.0.0
Code review is the last line of defense before code reaches production. This skill structures the review process to catch real issues — not just style preferences — and ensures every comment is actionable and proportionate.
1. Read the PR description fully — understand the intent before reading code. 2. Check: Does the implementation match the stated intent? 3. Identify the risk level: data mutation? auth changes? public API?
**Verify:** You understand what the PR is trying to accomplish.
4. Does the code do what it claims to do? 5. Are there off-by-one errors, null dereferences, or race conditions? 6. Are all error cases handled? 7. Do tests cover the happy path AND key failure paths?
**Verify:** You can trace the execution path for the primary use case and 2 failure cases.
8. Apply [security-hardening skill](../security-hardening/SKILL.md) to any auth/input/data changes. 9. Does this change open any OWASP Top 10 vulnerabilities? 10. Are any secrets or PII handled correctly?
11. Will the next developer understand this code without the author present? 12. Are functions doing one thing? 13. Are names descriptive and accurate? 14. Is complexity proportionate to the problem?
15. Every comment must be one of:
16. Blockers must be specific: *"This SQL query is vulnerable to injection via `{username}` — use parameterized queries."* 17. Never leave vague comments like *"this doesn't look right"* without explaining why.
| Excuse | Rebuttal | |--------|----------| | "I'll review it quickly" | A rushed review is not a review. Take the time or ask someone who can. | | "The tests pass so it's fine" | Tests prove the code works for tested inputs, not that it's secure or maintainable. | | "I'll comment on style later" | Style comments without blocker separation waste everyone's time. Label them. |
AI agent skills for production grade applications
Validates, parses, and sanitizes AI-generated outputs before they reach end users or downstream systems. Structured output enforcement, schema validation, and…
Design stable, versioned, self-documenting APIs. Easy to use correctly, hard to use incorrectly. Apply Hyrum's Law from day one.
Automated quality gates from commit to production. Every merge to main is potentially shippable. No manual steps in the deployment path.
Get layered, context-aware explanations of unfamiliar code. Understand what it does, why it was written that way, and how to work with it safely.
Load minimum necessary context into agent context windows. Prevents token bloat, reduces cost, and improves focus. Only load what the current task needs.
Systematic root cause analysis for production and development bugs. Hypothesis-driven debugging — never guess-and-check.