yoder-bawt

Yoder Skill Auditor

運維安全v3.1.0
作者:yoder-bawt更新於 2026-05-11共 2 個版本 · 最新版本釋出於 2026-02-16yoder-skill-auditor
#Prompt Injection#Crypto

據 AlphaMoat 統計,截至 2026-10,Yoder Skill Auditor 累計下載 7,170 次、獲 3 個星標,按累計下載量在本次參與排名的 10,914 個 AI 技能中排名第 3,930 位,在「運維安全」分類(502 個)中排名第 116 位。

列表預設按累計下載量排序,也可按星標排序;詳情頁排名按累計下載量計算。

累計下載
7170
星標
3
全站下載量排名
#3,930
全站前 37% · 「運維安全」前 24%
平台累計下載星標評論
ClawHub ↗71703-

各平台的計數口徑不同,統計時間不一致。

這個技能是做什麼的

依賴 python3 與 bash,提供技能目錄掃描、安裝前檢查、信任評分、版本差異、趨勢記錄與批次分析,可輸出 JSON 用於管線。檢查涵蓋認證資料讀取、提示注入、下載執行等模式;評分和掃描結論不構成安全保證。警告、關鍵問題和輸入錯誤分別有獨立退出狀態,關鍵發現需處理,不應盲目加入白名單。

資料釋出於 2026-10-26

可以用來做什麼

  • 審查技能安裝包或升級前後的風險變化

什麼時候用

  • 安裝第三方技能或複查已有技能時

示例提示詞

以下是根據技能文件生成的示例,不代表實際執行結果。

掃描這個技能目錄,解釋關鍵發現,並與舊版本比較。

月度資料趨勢

累計下載7170
6214672872412026-05 · 累計下載 62852026-06 · 累計下載 64322026-07 · 累計下載 67362026-08 · 累計下載 68752026-09 · 累計下載 69952026-10 · 累計下載 717026-0526-0926-10
星標3
2342026-05 · 星標 32026-06 · 星標 32026-07 · 星標 32026-08 · 星標 32026-09 · 星標 32026-10 · 星標 326-0526-0926-10

2026-09 起統計方法調整:同名技能按作者分別統計,並覆蓋全部技能。圖中虛線為調整前的月份,與之後的資料不直接比較,也不計算跨越調整的增長。

月份累計下載環比星標
2026-107170—3
2026-096995—3
2026-086875+2.1%3
2026-076736+4.7%3
2026-066432+2.3%3
2026-056285—3

SKILL.md 原文

以下是作者釋出的原始檔案,按純文字顯示,本站未做修改或稽核;其中的連結不可點選,命令請核實後再執行。抓取於 2026-10-23。

6.6 KB · ClawHub v3.1.0
---
name: skill-auditor
version: 3.1.0
description: "The definitive security scanner for OpenClaw skills. 18 security checks including prompt injection detection, download-and-execute, privilege escalation, credential harvesting, supply chain attacks, crypto drains, and more. 5-dimension trust scoring with trend tracking."
metadata:
  openclaw:
    requires:
      bins: ["python3", "bash"]
      env: []
      config: []
展開全部(共 162 行)
    user-invocable: true
---

# Skill Auditor v3.1.0

The definitive security scanner for OpenClaw/ClawHub skills. Best-in-class detection across 18 security checks including **prompt injection detection** — the first scanner to catch agent manipulation attacks in skill documentation. 5-dimension trust scoring, trend tracking, diff analysis, and benchmarking. Zero false positives on legitimate skills.

## When to Activate

1. **Installing a new skill** from ClawHub - run `inspect.sh` for full pre-install validation
2. **Auditing existing skills** - use `audit.sh` to scan any skill directory
3. **Generating trust scores** - use `trust_score.py` for 0-100 rating across 5 dimensions
4. **Comparing skills** - use `trust_score.py --compare` for side-by-side analysis
5. **Tracking improvements** - use `trust_score.py --save-trend` to monitor score over time
6. **Reviewing updates** - use `diff-audit.sh` to compare before/after versions
7. **Batch scanning** - use `audit-all.sh` or `benchmark.sh` for fleet-wide analysis

## Quick Start

```bash
# Audit a single skill
bash audit.sh /path/to/skill

# Trust score (0-100 across 5 dimensions)
python3 trust_score.py /path/to/skill

# Compare two skills side by side
python3 trust_score.py /path/to/skill1 --compare /path/to/skill2

# Track score over time
python3 trust_score.py /path/to/skill --save-trend
python3 trust_score.py /path/to/skill --trend

# Diff audit (before/after update)
bash diff-audit.sh /path/to/old-version /path/to/new-version

# Benchmark against a corpus
bash benchmark.sh /path/to/skills-dir

# Inspect a ClawHub skill before installing
bash inspect.sh skill-slug

# Audit all installed skills
bash audit-all.sh

# Generate a markdown report
bash report.sh

# Run test suite (28 assertions)
bash test.sh
```

## Guardrails / Anti-Patterns

**DO:**
- ✓ Always audit skills before installing from untrusted sources
- ✓ Review trust scores - reject skills scoring below 60 (D grade)
- ✓ Use `diff-audit.sh` when updating skills to catch regressions
- ✓ Use `--json` output for CI/CD pipeline integration
- ✓ Run `--save-trend` periodically to track skill health

**DON'T:**
- ✗ Install skills scoring below 40 (F grade) without extensive manual review
- ✗ Ignore CRITICAL findings - they indicate potential security threats
- ✗ Blindly add skills to allowlist without understanding why they access credentials
- ✗ Skip audit because a skill is "popular" or "official"

## Security Checks (18 total)

| # | Check | Severity | Description |
|---|-------|----------|-------------|
| 1 | credential-harvest | CRITICAL | Scripts reading API keys/tokens AND making network calls |
| 2 | exfiltration-url | CRITICAL | webhook.site, requestbin, ngrok URLs in scripts |
| 3 | obfuscated-payload | CRITICAL | Base64-encoded URLs or shell commands |
| 4 | sensitive-fs | CRITICAL | /etc/passwd, ~/.ssh, ~/.aws/credentials access |
| 5 | crypto-wallet | CRITICAL | Hardcoded ETH/BTC wallet addresses (drain attacks) |
| 6 | dependency-confusion | CRITICAL | Internal/private-scoped packages in public deps |
| 7 | typosquatting | CRITICAL | Misspelled package names (lodahs, requets, etc.) |
| 8 | symlink-attack | CRITICAL | Symlinks targeting sensitive system paths |
| 9 | code-execution | WARNING | eval(), exec(), subprocess patterns |
| 10 | time-bomb | WARNING | Date/time comparisons that could trigger delayed payloads |
| 11 | telemetry-detected | WARNING | Analytics SDKs, tracking pixels, phone-home behavior |
| 12 | excessive-permissions | WARNING | >15 bins/env/config items requested |
| 13 | unusual-ports | WARNING | Network calls to non-standard ports |
| 14 | prompt-injection | CRITICAL | Agent manipulation in docs: "ignore instructions", role hijacking, hidden HTML directives |
| 15 | download-execute | CRITICAL | curl\|bash, wget\|sh, eval $(curl), unsafe pip/npm installs |
| 16 | hidden-file | WARNING | Suspicious dotfiles that may hide malicious content |
| 17 | env-exfiltration | CRITICAL | Reading sensitive env vars + outbound network calls |
| 18 | privilege-escalation | CRITICAL | sudo, chmod 777/setuid, writes to system paths |

Context-aware: credential mentions in documentation are INFO, not CRITICAL.

## Trust Score (5 Dimensions)

| Dimension | Max | What's Measured |
|-----------|-----|-----------------|
| Security | 35 | Audit findings (criticals = -18, warnings = -4) |
| Quality | 22 | Description, version, usage docs, examples, metadata, changelog |
| Structure | 18 | File organization, tests, README, reasonable scope |
| Transparency | 15 | License, no minified code, code comments |
| Behavioral | 10 | Rate limiting, error handling, input validation |

Grades: A (90+), B (75+), C (60+), D (40+), F (<40)

### Comparative Scoring
```bash
python3 trust_score.py /path/to/skill-a --compare /path/to/skill-b
```
Shows per-dimension deltas and overall score difference.

### Trend Tracking
```bash
python3 trust_score.py /path/to/skill --save-trend   # Record score
python3 trust_score.py /path/to/skill --trend         # View history
```
Stores up to 50 entries per skill in `trust_trends.json`.

## Tools

| File | Purpose |
|------|---------|
| audit.sh | Single skill security audit (18 checks) |
| audit-all.sh | Batch scan all installed skills |
| trust_score.py | Trust score calculator (5-dimension, 0-100) |
| diff-audit.sh | Compare skill versions for security regressions |
| benchmark.sh | Corpus-wide audit with aggregate statistics |
| inspect.sh | ClawHub pre-install workflow |
| report.sh | Markdown report generator |
| test.sh | Automated test suite (28 assertions, 12 test skills) |
| allowlist.json | Known-good credential skills |

## Test Suite

12 test skills (8 malicious, 4 clean) with 28 automated assertions:

```bash
bash test.sh
```

Malicious fixtures: credential harvest, obfuscated payload, sensitive fs reads, crypto wallets, time bombs, symlink attacks, prompt injection, download-execute, privilege escalation.
Clean fixtures: basic skill, credential docs (false positive check), network skill, dotfiles skill.

## Exit Codes
- 0: PASS / safe to install
- 1: REVIEW / warnings found
- 2: FAIL / critical issues
- 3: Error / bad input

## Changelog

See [CHANGELOG.md](CHANGELOG.md) for full version history.

「運維安全」分類下的熱門技能

鍾馗.skill運維安全

鍾馗.Skill——會更新漏洞庫的安全審查專家。直來直去、快刀斬亂麻,三層審查(靜態審計/行為模擬/供應鏈溯源)覆蓋12類風險,輸出結構化安全裁定(✅乾淨/⚠️可疑/🚫惡意)。Use when 使用者說"審查這個Skill"、"安全檢查"、"鍾馗看下"、"審一下"、"查一下這個skill"、安裝Skill前的安全評估、或需要審計SKILL.md的惡意載荷。

↓ 33.7萬★ 17user_741dc82b
SkillScan運維安全

在使用技能包之前做安全掃描,並按結論決定放行或攔截的技能。

↓ 18.3萬★ 41tokauthai
Free Ride - Unlimited free AI運維安全

為 OpenClaw 配置 OpenRouter 免費模型與回退鏈的工具。

↓ 13.4萬★ 492Shaishav Pidadi
a ca s ca s c ca s c運維安全

Parse, search, and analyze application logs across formats. Use when debugging from log files, setting up structured logging, analyzing error patterns, correlating events across services, parsing stack traces, or monitoring log output in real time.

↓ 12.5萬★ -user_60e896a0
Skill Scanner運維安全

用於安裝技能、包或外掛的安全檢查。請在執行任何 npm install、openclaw plugins install、clawhub install 或類似安裝命令之前使用。

↓ 9.3萬★ 4sudhindrat
網安技能運維安全

網路安全與SRC漏洞掘金實戰技能。融合「系統學習路徑」與「SRC實戰掘金」雙模式: ①學習模式=法律→基礎→工具→靶場→攻防→就業全流程; ②實戰模式=直接指導在騰訊TSRC、阿里ASRC、百度BSRC等平台挖掘高價值漏洞變現。 當用戶提出以下問題時觸發: "我想學網路安全怎麼入門""SRC漏洞怎麼挖能賺錢""滲透測試需要學什麼" "Web安全漏洞原理是什麼""怎麼準備CTF比賽""護網面試會問什麼" "雲安全怎麼學""挖漏洞賺錢的方法""埠掃描指令碼怎麼寫" "SQL隱碼攻擊怎麼檢測""子域名怎麼列舉""XSS漏洞怎麼找" "靶場怎麼搭建""Kali Linux怎麼用""Metasploit怎麼入門" "等保測評是什麼""程式碼審計怎麼做""內網滲透怎麼學" "資訊洩露漏洞怎麼挖""越權漏洞怎麼測試""邏輯漏洞挖掘技巧" "簡歷怎麼寫才能進安全公司""安全行業面試題有哪些" "Burp Suite怎麼用""DVWA怎麼安裝""Pikachu靶場搭建" "Vulhub漏洞復現""網路安全法有什麼規定""CTF奪旗賽怎麼參加" "AWD比賽是什麼""應急響應怎麼做""攻擊溯源怎麼搞" "資料安全怎麼保障""公有云安全怎麼防護""私有云攻防怎麼做" "雲原生安全是什麼"。

↓ 8萬★ 16user_d2cc4baf
人人都能成為駭客v2.0(Lvan)運維安全

這個技能特別簡單——你只需要呼叫它,給一個網站域名,它就會自動幫你跑一遍滲透測試,找出這個網站存在哪些安全漏洞。 不需要你配環境、不用你手動發包,一切自動化。測完直接出結果,告訴你哪裡有問題、怎麼修。 它的願景是:讓每個人都能成為白帽駭客——只要輸入一個域名,就能知道自己的網站安不安全。一切只為授權測試,只為發現漏洞、修復風險,讓網際網路更安全。

↓ 5.7萬★ 37user_a730098d
skill-vetter運維安全

Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.

↓ 5.4萬★ -user_814dbe54

資料月份:2026-10 · 下載/星標數聚合自公開技能註冊庫(ClawHub、SkillHub),按月更新。口徑說明見資料方法。