security-audit SKILL.md skillA

security-audit from huanmin123/ai-vibecode-superpower is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add huanmin123/ai-vibecode-superpower/security-audit@git:20261104.d6f23e3

Pins git:20261104.d6f23e3 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
huanmin123/ai-vibecode-superpower
Path
shared/skills/security-audit/SKILL.md
Versions recorded
2
Latest version
git:20261104.d6f23e3, committed Nov 4, 2026
First version recorded
Nov 4, 2026
Size
16,907 bytes
Repository stars
39
License
Apache-2.0

对目标项目及其运行环境做取证式安全审计:先测绘攻击面,再以攻击者视角建立攻击路径,按通用视角做多视角穷举排查,逐条用适用层的证据证实或证伪,并对外部漏洞事实做可追溯核实,输出含覆盖台账的审计报告。用于安全审计、漏洞排查、上线前体检、PR 安全审查、部署/主机/云配置审查;不用于依赖升级、普通功能开发、单点 bug 修复或未授权渗透测试。.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# 安全审计

审计交付的是**可核对的结论**,不是漏洞清单。两条主线:

1. **攻击者视角**:从"谁、已经有什么、想得到什么、最便宜的路径是什么"组织排查,而不是从漏洞名出发。
2. **外部核实**:任何版本默认值、已知漏洞、修复版本、在野利用,都必须追溯到审计时打开过的来源。

> 本文所有入口类型、视角、层、画像、漏洞类、工具与来源列表都是**下界,不是全集**。看到任何清单,先问"这个目标还有哪些表外情况",再按清单展开;不得因为"表里没有"而跳过。

## 定位与边界

| 任务形态 | 走向 |
| --- | --- |
| 源码、配置、供应链、部署、主机、云、CI/CD 安全审查 | 本 skill |
| 质量缺陷、隐藏 bug、回归风险 | `deep-dive-review` |
| 跨文件实现、修复与独立验收 | `orchestrate-model-workflow` |
| 依赖升级、普通功能开发、单点 bug 修复 | 不进本 skill |
| 未授权渗透、主动打靶、红队 | 不在本 skill 范围 |

默认**只读**:不修改目标仓库的源码与配置,不对目标系统发起攻击性请求。报告写到用户指定路径;未指定时用项目外临时目录或聊天正文,不写入目标仓库源码/配置。需要修复时由用户另行发起;需要运行时/主机/云验证时,必须有明确授权并限定范围。

## 不变量

违反任一条,该结论作废。

1. **攻击者目标优先**:每条 finding 要能回答"哪个画像、已有什么、想得到什么、还差哪一步"。只写"这里有个危险调用"不构成 finding。
2. **证据准入**:代码 finding 必须有 `source(file:line) → 传播路径 → sink(file:line)`,每一跳实际读取核对。非代码 finding 必须有适用层的可复核证据链:定位授权资源或只读命令、执行身份与时间、相关输出及控制和影响;区分已观察事实与未验证的部署或运行条件。只有模式匹配的,登记为线索。
3. **控制判定**:区分"控制缺失"与"控制存在但无效",指出具体位置。框架/库默认行为必须按安装版本核实,不凭记忆。
4. **层与证据等级**:结论要标注属于哪一层(代码/配置/供应链/主机/网络/云/CI/数据)与证据等级(E1 仓库、E2 部署产物、E3 只读运行访问、E4 文档访谈)。不得把 E1/E2 推断写成 E3 事实。
5. **外部事实可追溯**:CVE/GHSA/公告/修复版本/默认行为必须给来源、版本、检索日期;没打开过的来源不引用,不编造编号或引文。
6. **时效性**:安全事实会过期。示例与历史结论只是起点,审计时重新核实;无法核实的写"未验证"。
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add huanmin123/ai-vibecode-superpower/security-audit@sha256:171037f49e240eaf

The line at the top of this page pins the label git:20261104.d6f23e3, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_zrukh6kcg4pf5hjp.svg)](https://markdownregistry.com/a/art_zrukh6kcg4pf5hjp)

0 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 2 versions of this file, all committed on Nov 4, 2026.

versioncommittedcommitsizeaudit
git:20261104.d6f23e3 latest2026-11-04 d6f23e3 16,907 BA view · diff
git:20261104.29776112026-11-04 2977611 15,923 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (16907 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

huanmin123/ai-vibecode-superpower · 39 stars · license Apache-2.0 · pushed 2026-11-06 · branch master

API

GET https://markdownregistry.com/api/v1/artifacts/art_zrukh6kcg4pf5hjp
GET https://markdownregistry.com/api/v1/resolve?ref=huanmin123/ai-vibecode-superpower/security-audit
GET https://markdownregistry.com/api/v1/blob/171037f49e240eaffdd98b49a7c5e052337a02b6e0a68054f899c098d18c6d5d

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from huanmin123/ai-vibecode-superpower

AGENTS.md AGENTS.md file
huanmin123/ai-vibecode-superpower · AGENTS.md
git:20261106.ba19f6c · audit A · 39 stars
agent-toolchain SKILL.md skill
huanmin123/ai-vibecode-superpower · shared/skills/agent-toolchain/SKILL.md · 为目标项目受控接入、升级、修复或维护 CodeGraph 与 RTK;在写入项目配置、安装受管版本、重建索引、健康检查或回滚时使用。
git:20261106.ba19f6c · audit A · 39 stars
deep-dive-review SKILL.md skill
huanmin123/ai-vibecode-superpower · shared/skills/deep-dive-review/SKILL.md · 对已交付或即将交付的需求、模块或变更集做多轮深挖质量审查:按覆盖矩阵开展假设穷举、对抗证伪、裁决过滤及证据链扩散,并按模式执行终态复核与收敛判据。显式发起即默认 report-only;只有用户显式确认 A 类即修权后才进入 fix…
git:20261104.d0edca9 · audit A · 39 stars
orchestrate-model-workflow SKILL.md skill
huanmin123/ai-vibecode-superpower · shared/skills/orchestrate-model-workflow/SKILL.md · 用于复杂开发、跨文件实现或修复、架构设计、Bug 或漏洞诊断、复审和验收;按证据、风险和执行边界选择角色,可拆任务优先多子代理并行。
git:20261104.becdd62 · audit A · 39 stars
project-doc-planner SKILL.md skill
huanmin123/ai-vibecode-superpower · shared/skills/project-doc-planner/SKILL.md · 设计或更新项目文档架构、根目录 AGENTS.md、开发规范、决策记录与可复用模板,以及 `.local` 私有层、私有环境资源与交付流程。用于文档规划或整理,不因普通开发、部署或修复任务自动重建文档体系。
git:20261103.7c25c4b · audit A · 39 stars
test-strategy SKILL.md skill
huanmin123/ai-vibecode-superpower · shared/skills/test-strategy/SKILL.md · 用于制定或调整测试策略、选择测试范围与命令、决定是否跑全量/场景/性能测试;普通编码沿用项目既有测试命令,不自动改写测试结构。
git:20261104.d6f23e3 · audit A · 39 stars
visual-aesthetics SKILL.md skill
huanmin123/ai-vibecode-superpower · shared/skills/visual-aesthetics/SKILL.md · 当视觉呈现是任务目标、改动对象或验收条件时使用。按形式、内容与情境之间恰当而有表现力的关系组织或评审视觉产出,并以实际呈现为证据。纯功能改动、普通文本答复或没有视觉呈现的日志不触发。
git:20261021.95b28f7 · audit A · 39 stars

Every file in huanmin123/ai-vibecode-superpower

Other files named security-audit

security-audit SKILL.md skill
davepoon/buildwithclaude · plugins/agent-triforce/skills/security-audit/SKILL.md · Deep security audit covering OWASP Top 10, authentication, authorization, data protection, dependency vulnerabilities…
git:20260307.756e327 · audit A · 3,724 stars
security-audit SKILL.md skill
maxmiksa/auto-company · .claude/skills/security-audit/SKILL.md · Use when reviewing code security, auditing dependencies for CVEs, checking configuration or secret security, assessing…
git:20250919.3118c12 · audit A · 3,191 stars
security-audit SKILL.md skill
borghei/claude-skills · engineering/security-audit/SKILL.md · Multi-agent source security audit with a coverage ledger, independently verified findings, and machine-checked records…
v1.0.0 · audit A · 1,075 stars
security-audit SKILL.md skill
opencrabs/opencrabs · src/docs/reference/templates/skills/security-audit/SKILL.md · Run a comprehensive language-agnostic security & CVE audit, scored 0-100
git:20260501.09a3f3d · audit A · 1,027 stars
security-audit SKILL.md skill
laolaoshiren/claude-code-skills-zh · skills/security-audit/SKILL.md · 代码安全审计 - 漏洞扫描、依赖检查、安全最佳实践
git:20260421.27394be · audit A · 869 stars
security-audit SKILL.md skill
decebals/claude-code-java · skills/security-audit/SKILL.md · Java security checklist covering OWASP Top 10, input validation, injection prevention, and secure coding. Works with…
git:20260828.702770a · audit B · 768 stars
security-audit SKILL.md skill
cowork-os/cowork-os · resources/skills/security-audit/SKILL.md · Answer security questions about code, review a file or change for vulnerabilities, or run a full multi-agent audit of a…
v2.0.0 · audit A · 535 stars
security-audit SKILL.md skill
josstei/maestro-orchestrate · plugins/maestro/skills/security-audit/SKILL.md · Run a Maestro-style security assessment for authentication, authorization, data exposure, secret handling, and…
git:20260410.f5a5165 · audit A · 486 stars

Browse by kind, by grade A, or by owner.