deep-review

仓库创建 2026年6月26日最近提交 22 天前SkillHot 收录 22 天前

▸ 精选理由

适用于发布前审计与发现系统性问题。

▸ 风险提示

需要访问完整仓库，可能触及敏感代码或凭据

这个 Skill 做什么

对现有代码库与设计进行全面、深入的质量与债务审核。

对现有代码库和设计做一次全面而严格的审计，找出技术债、重复、规范漂移和需要优先修复的问题。适合在发布前审查、评估项目健康或有人问“哪里能做得更好”的场景。它会并行展开不同维度的检查，自动修复明显问题并把需要人为判断的点清晰列出，避免泛泛而谈。

▸ 展开 SKILL.md 英文原文

Run a thorough, high-standard review of the EXISTING codebase and design (not a new feature). Use when asked to "review the whole solution", "audit the project before release", find "what can be done better", or assess technical debt, duplication, or convention drift in depth. Fans out parallel dimension audits, fixes what is unambiguously right, and surfaces the judgment calls.

开发编程代码审计技术债务全面评估审查通用

Stars

Forks

仓库内 Skill

7 日增星

安装 / 使用

给你的 Agent 一句话（通用）

帮我安装这个 skill：https://raw.githubusercontent.com/Svendsys/agent-skills/master/skills/deep-review/SKILL.md

或 curl 直取 SKILL.md

curl -fsSL "https://raw.githubusercontent.com/Svendsys/agent-skills/master/skills/deep-review/SKILL.md"

SKILL.MD 节选查看完整文件 ↗

# Deep Review

Repository-wide quality review: parallel dimension audits, a hard
safe-vs-needs-input split, re-verify before acting. Breadth comes from fanning
out; trust comes from the orchestrator validating every finding against source.

## 1. Ground the review

Read `AGENTS.md` / `CLAUDE.md` and any conventions docs — **these are the bar**;
measure against the project's own standards, not generic best practice. Capture
a baseline so findings aren't vibes: build/test time, output size and largest
artifacts, suite green. Note the numbers.

## 2. Fan out one audit per dimension (parallel, read-only)

Three dimensions apply to any project:

1. **Conventions / duplication / single source of t

via SKILL·HOT · 数据来自 GitHub 公开信息 · 原文版权归作者所有