Skills for Real Engineers. Straight from my .agents directory.
code-review-patterns
Agent skill froma5c-ai/babysitter
Multi-dimensional code assessment across security, quality, performance, and maintainability with confidence-gated reporting (>=80%) and Router Contract generation.
How to install code-review-patterns
npx skills add a5c-ai/babysitter# or add only this skill to Claude Code: copy the folder
library/methodologies/cc10x/skills/code-review-patterns/
# into
~/.claude/skills/code-review-patterns/ (all projects)
.claude/skills/code-review-patterns/ (one project)
1- Injection vulnerabilities (SQL, XSS, command)2- Authentication and authorization patterns3- Secrets handling (no hardcoded credentials)4- Input validation and sanitization56### Quality (weight: 25%)7- Naming conventions and consistency8- Code structure (SRP, cohesion, coupling)9- Error handling completeness10- Type safety (no `any` escapes)1112### Performance (weight: 20%)13- Algorithmic complexity (flag O(n^2) or worse)14- Resource and memory leaks15- Database query efficiency (N+1)16- Caching opportunities1718### Maintainability (weight: 25%)19- Documentation (JSDoc/TSDoc for public APIs)20- Test coverage adequacy21- Readability (function length, nesting depth)22- Tech debt markers (TODO, FIXME)2324## Confidence Gating2526- Only report issues with confidence >= 80%27- Empty catch blocks are always critical (100% confidence)28- Classify: critical, high, medium, low29- Include actionable remediation for each issue3031## Router Contract3233Every review must produce: STATUS, BLOCKING, REQUIRES_REMEDIATION, issue counts.3435## When to Use3637- Code review step in BUILD workflow38- Fix review in DEBUG workflow39- Full REVIEW workflow4041## Agents Used4243- `code-reviewer` (primary consumer)44- `silent-failure-hunter` (error handling dimension)
Excerpt shown under the repository's MIT license.
More skills in babysitter
Similar skill repositories
An agentic skills framework & software development methodology that works.
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
Public repository for Agent Skills
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
