用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt;运行前请先检查来源。
npx skills add https://github.com/JSai23/claude-tooling --skill verify-style-a命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地?可下载 SkillsMP 当前能够提供的文件。
Write high-quality session prompts (SESSION_WORKER.md and SESSION_REVIEWER.md) for an agent loop. Reads repo context and follows the loop-author guide to produce targeted, effective prompts. Use when setting up a new loop or rewriting session prompts for an existing one.
Pre-flight checks and launch for agent loops. Validates branch state, session prompts, creates loop directory structure, and launches run.sh in a tmux session. Use when the user wants to start a new agent loop in the current repo.
Cross-repo visibility for agent loops. Lists running, completed, and crashed loops. Reads the registry and per-repo tracking logs. Shows tmux attach commands. Use when checking what loops are active, what ran recently, or loop details.
正在显示 SKILL.md
基于 SOC 职业分类
| name | verify-style-a |
| type | action |
| description | Code cleanliness — AI-generated slop, naming, dead code, over-engineering, simplification |
| user-invocable | false |
Action skill — Code cleanliness review: AI slop detection, naming, dead code, fix protocol, output format.
Line-level code quality without changing behavior.
Defensive code — Try/catch around code that can't fail. Null checks on values that are never null. Fallbacks for impossible cases. Validation of internal values guaranteed by the type system.
Over-engineering — Abstractions for one-time operations. Interfaces with single implementations. Factory patterns for simple construction. Configuration for things that will never be configured.
Verbosity — Multiple lines that could be one. Intermediate variables that add nothing. Obvious comments that restate the code.
Naming — Unclear names. Inconsistent naming across similar modules. Abbreviations that aren't obvious. Names that lie.
Structure — Functions doing multiple things. Files with mixed responsibilities. Duplicated logic. Complex conditionals that could be simplified.
Dead code — Unused imports. Unreachable branches. Functions nothing calls. Commented-out code.
Find, fix, test. One issue at a time. If tests fail, revert and note why.
Not architecture review, not larp detection. Don't change behavior. "No issues found" is valid.
## Fixes Applied
### 1. {path}:{line} - {category}
BEFORE: {old code}
AFTER: {new code}
REASON: Why this was changed
## Skipped (tests failed)
- {path}:{line} - what you tried, why it broke tests
## Summary
- Fixed: N issues
- Skipped: M issues (would break tests)
- Lines removed: K