菜单
Unified verification engine for Python data science projects. Covers environment checks, type checking, linting, tests, security scans, code review with DS anti-patterns, and notebook checks. Commands (/verify, /quality-gate) invoke different subsets of this skill.
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。
基于 SOC 职业分类
| name | verification-loop |
| version | 1.0 |
| description | Unified verification engine for Python data science projects. Covers environment checks, type checking, linting, tests, security scans, code review with DS anti-patterns, and notebook checks. Commands (/verify, /quality-gate) invoke different subsets of this skill. |
The single source of truth for all code checks. Commands invoke specific phases:
/verify → Phases 1-4 (does my code work?)/quality-gate → Phases 1-4 + Phase 6 (safe to push?)Phase 5 (code review) runs when the user asks for a code review directly.
Read skills/phases.md for all 7 phases with exact commands, the code review checklist, and the output format.
Read guidelines.md for verdict rules, coverage targets, and escalation criteria.
Use when writing Python error handling code. Provides specific rules for exception chaining and custom error hierarchies.
Use when reviewing code for security issues.
Use when implementing features, fixing bugs, or refactoring code that may invalidate existing docs. Detects stale documentation by matching the code diff against in-repo doc files and applies targeted updates. Relevant when code changes rename, remove, or add APIs, fields, config keys, or CLI flags. Also applies when the user says "update docs", "check docs", or "are the docs stale", or when reviewing a branch for documentation accuracy.
Helps you write better Python code by following clean code principles and software engineering best practices.
Use when the user asks to design, plan, or explore approaches before implementing — creating features, building components, or adding functionality that would benefit from design exploration first.
Measurement-driven code refactoring — profile before changing, measure after, keep only if metrics improve. Covers complexity reduction, extraction patterns, and bulk refactoring for mechanical changes across many files.