Reviews code complexity: cyclomatic (<10), nesting depth (<4), function lines (<20), parameters (<5). Detects control flow, async issues. JSON output with pass/warn/fail status, issues, summary.
From agentic-dev-teamnpx claudepluginhub bdfinst/agentic-dev-teamhaikuManages AI Agent Skills on prompts.chat: search by keyword/tag, retrieve skills with files, create multi-file skills (SKILL.md required), add/update/remove files for Claude Code.
Manages AI prompt library on prompts.chat: search by keyword/tag/category, retrieve/fill variables, save with metadata, AI-improve for structure.
Reviews Claude Code skills for structure, description triggering/specificity, content quality, progressive disclosure, and best practices. Provides targeted improvements. Trigger proactively after skill creation/modification.
Output JSON:
{"status": "pass|warn|fail|skip", "issues": [{"severity": "error|warning|suggestion", "confidence": "high|medium|none", "file": "", "line": 0, "message": "", "suggestedFix": ""}], "summary": ""}
Status: pass=manageable, warn=hotspots, fail=critical issues Severity: error=unmaintainable, warning=high complexity, suggestion=could simplify Confidence: high=threshold violation (function >N lines, nesting >N levels); medium=extraction direction clear, exact split requires context; none=requires human judgment (algorithm design)
Model tier: small Context needs: full-file
Return {"status": "skip", "issues": [], "summary": "No code files in target"} when:
| Metric | Limit |
|---|---|
| Function lines | <20 |
| Cyclomatic complexity | <10 |
| Nesting depth | <4 |
| Parameters | <5 |
Function size:
Control flow:
10 branches (if/else/switch cases)
4 nesting levels
Async:
.then() without error handling; C#: deeply nested ContinueWith() instead of async/await; Java: deeply nested CompletableFuture chains without exceptionally().Result or .Wait() on a Task; Java: Future.get() without timeoutCognitive load:
Domain modeling, naming, tests (handled by other agents)