Principles Keep it simple (KISS) — prefer the simplest solution that works Don't repeat yourself (DRY) — extract when you see three duplicates, not before Single Responsibility — each module/function does one thing Use descriptive, intention-revealing names Use kebab-case for files and folders Functions should have clear inputs and outputs with minimal side effects Keep functions right-sized — extract when logic needs a comment to explain Delete dead code — don't comment it out Never swallow errors silently Measure before optimizing — no premature performance work No premature abstraction — wait for three concrete duplicates before extracting Rules See rules index for detailed patterns. Examples Positive Trigger User: "Review this service and remove duplication while keeping behavior unchanged." Expected behavior: Use core-coding-standards guidance, follow its workflow, and return actionable output. Non-Trigger User: "Generate a one-off product marketing tagline." Expected behavior: Do not prioritize core-coding-standards ; choose a more relevant skill or proceed without it. Troubleshooting Skill Does Not Trigger Error: The skill is not selected when expected. Cause: Request wording does not clearly match the description trigger conditions. Solution: Rephrase with explicit domain/task keywords from the description and retry. Guidance Conflicts With Another Skill Error: Instructions from multiple skills conflict in one task. Cause: Overlapping scope across loaded skills. Solution: State which skill is authoritative for the current step and apply that workflow first. Output Is Too Generic Error: Result lacks concrete, actionable detail. Cause: Task input omitted context, constraints, or target format. Solution: Add specific constraints (environment, scope, format, success criteria) and rerun. Workflow Identify whether the request clearly matches core-coding-standards scope and triggers. Apply the skill rules and referenced guidance to produce a concrete result. Validate output quality against constraints; if gaps remain, refine once with explicit assumptions.

安装