audit

仓库创建 2026年7月2日最近提交 22 天前SkillHot 收录 22 天前
▸ 精选理由

端到端的自动审计与修复流程,适合需要系统安全与质量把关的团队。

▸ 风险提示

可能需要对代码库写权限与高权限操作,具备变更风险。

这个 Skill 做什么

按顺序运行完整代码库审计链,产出并可选择实现修复的综合报告。

把整个代码库从架构到安全到工具链完整地跑一遍审计流程:先检查结构、再查安全、再核对工具和机器/人工文档,必要时按顺序派出子团队去实现修复并产出综合报告。遇到要全面把控风险或在大改之前做健康检查就用。特点是按顺序用多个 agent 分工审查,有权限的步骤还能直接落地修复。

▸ 展开 SKILL.md 英文原文

Run a full codebase audit chain — architecture, security, toolkit integrity, machine docs, human docs, then security again. Creates a team, spawns agents consecutively, and reports results.

开发编程代码审计安全自动化链通用
0
Stars
0
Forks
35
仓库内 Skill
+0
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/dustinkeeton/wafflestack/main/.claude/skills/audit/SKILL.md
或 curl 直取 SKILL.md
curl -fsSL "https://raw.githubusercontent.com/dustinkeeton/wafflestack/main/.claude/skills/audit/SKILL.md"
SKILL.MD 节选查看完整文件 ↗
# Codebase Audit Chain

Run the full audit pipeline in consecutive order. Each agent audits the codebase and (where its role grants edit tools) implements fixes before the next one starts; report-only agents deliver findings for the user or a later agent to fix.

## Chain Order

1. **general-purpose** — Audit and improve codebase structure (module patterns, file organization, naming, dependency rules, import paths)
2. **general-purpose** (pass 1) — Full security audit per the security-audit skill checklist
3. **general-purpose** (Toolkit integrity) — Validates manifests, placeholders, tests, and render/lock integrity (validate + test + render + doctor).
4. **docs-agent** — Create/update the 
via SKILL·HOT · 数据来自 GitHub 公开信息 · 原文版权归作者所有