code-quality
Review checklist for Java / Spring Boot code - correctness, security, performance, Spring signals, REST semantics, and code smells; project rules are referenced from conventions, not repeated. Use when user says "review code", "check this PR", "refactor", "check API", or before merging changes.
92%
Does it follow best practices?
Run evals on this skill
Adds up to 20 points to the overall score
Passed
No findings from the security scan
Code Quality Review Skill
Knowledge for reviewing Java / Spring Boot changes: what to flag and how to recognize it. The review process, scope, and report format belong to the code-reviewer agent; project rules belong to the conventions skill.
No duplication: this skill holds only what no other skill states (generic Java quality, smells, REST semantics, detection signals). Project rules live in conventions, spring-boot, and jpa-patterns; the references here point to them and never restate them, so a rule change is made in one place.
When to Use
- "review this code" / "code review" / "check this PR"
- "refactor" / "clean this code" / "improve readability"
- "review API" / "check endpoints" / "REST review"
- Before merging a PR or releasing API changes
What to Load
Read only the references that match the changed files.
| If the change contains | Read |
|---|---|
| Any Java code | references/correctness.md, references/spring-conventions.md (signals; rules in conventions) |
| Controllers, DTOs, OpenAPI, error handling | references/api-contract.md |
| Input handling, auth, config, secrets, actuator | references/security.md |
| Repositories, queries, entity loops | references/performance.md |
Entities, *.sql migrations, spring.jpa.*, Flyway | conventions → Relational Database Schema (Flyway); spring-boot → references/data.md |
| Any non-trivial logic | references/smells.md |
Related skills, loaded only when the diff needs them:
| Topic | Skill |
|---|---|
| Project rules (No Lombok, RestClient, Flyway, Problem Details) | conventions |
| Lazy loading, fetching, optimistic locking | jpa-patterns |
Test quality, RestTestClient, Testcontainers | testing-patterns |
| Log levels, MDC, structured logging | logging-patterns |
| UI contract details | spring-boot → references/ui-contract.md |
| Records, sealed types, pattern matching | java-modern |
Severity
- Critical - security, data loss, crash risk → must fix before merge
- Important - correctness, performance, maintainability → should fix
- Code Smell - style, complexity, minor issues → nice to have (always phrased as a heuristic, see
references/smells.md) - Good - positive feedback worth reinforcing
Non-Goals
- Formatting, import order, naming style that a formatter or linter enforces
- Personal taste or rewrites that do not fix a concrete problem
- Rules the repo contradicts on purpose: when
CLAUDE.md,conventions, or an existing documented decision overrides a rule here, the repo wins
Quick Reference
| Category | Red flags |
|---|---|
| Null safety | chained calls, Optional.get(), returning null |
| Exceptions | empty catch, broad catch, lost stack trace |
| Resources | manual close() |
| Transactions | multi-step writes without @Transactional, self-invocation |
| API | wrong verb or status |
| Performance | N+1, findAll() without paging, LIKE from user input |
- Repository
piomin/claude-ai-spring-boot
- Revision
31fe1d5
- Last updated
- First committed
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.