Clean Code

**Judgement:** A strong, opinionated clean-code skill that is well-suited for production-oriented coding tasks and multi-agent workflows, but somewhat rigid and not ideal for educational or highly exploratory scenarios. **Strengths:** - Solid grounding in mainstream practices (SRP, DRY, KISS, YAGNI) applied through clear, concrete rules on naming, function size, arguments, and control flow. - Emphasizes disciplined workflow: think before editing, update all dependent files, avoid leaving broken imports, and run appropriate validation scripts per agent role. - Self-check list (goal met, files edited, tests/lint/types, edge cases) meaningfully improves reliability of delivered code. - Anti-pattern list (no pointless factories, utils files with one function, deep nesting, or comment spam) helps keep AI outputs pragmatic and lean. **Risks / Limitations:** - Strong bias against explanations and comments can conflict with users who explicitly want learning, documentation, or design rationale, and can reduce transparency into the model’s reasoning. - Rigid size and abstraction limits (e.g., max ~20 lines per function, minimal nesting) may be suboptimal for certain domains (e.g., numerics, DSLs, heavy framework glue) where slightly larger functions or more context are clearer. - Script-handling protocol (always summarize results, ask before fixing) improves control but can slow workflows when users actually prefer fully autonomous fixes. - Skill is language- and stack-agnostic, so it doesn’t encode any deep language-specific idioms or framework best practices. **Recommended Scenarios:** - Teams using OpenClaw for production coding, who value terse, maintainable code and want consistent structure and verification across agents. - Automated bug-fixing, refactoring, and incremental feature work in existing codebases, where dependency awareness and not breaking imports is critical. - Multi-agent pipelines where strict mapping of agents to their own QA scripts (lint, tests, security scans, SEO, etc.) and structured reporting of script results is desired. **Use with Caution:** - Educational contexts, tutorials, or onboarding material where detailed explanations and comments are required. - Design-heavy or research/prototyping phases where exploring alternatives and documenting trade-offs matters more than minimalism and brevity.