Skill Audit ๐
ๆซๆ OpenClaw skills ไธญ็ๅฎๅ จ้ฃ้ฉ๏ผ้ฒๆญขไพๅบ้พๆปๅปใ
About this skill
Skill Audit is a critical security tool designed for the OpenClaw AI agent ecosystem. Its primary function is to scan both installed skills and those available on ClawHub for potential security vulnerabilities and malicious code patterns. It categorizes risks into Critical (e.g., reading credential files, data exfiltration, arbitrary code execution), Warning (e.g., untrusted network requests, file system traversal), and Info (e.g., shell command usage, many dependencies) levels. Users can invoke the skill to scan all currently installed skills, target a specific skill by name or local directory, or proactively check a skill on ClawHub before installation. This systematic approach ensures that AI agents operate in a more secure environment, reducing the risk of compromise from third-party code. The skill provides clear, actionable reports, including a risk score and specific findings, aiding developers and users in making informed decisions about skill trustworthiness. By integrating Skill Audit into their workflow, users can protect their sensitive data, prevent system integrity breaches, and mitigate the threat of supply chain attacks inherent in using external code components. It acts as an essential gatekeeper, offering peace of mind to anyone leveraging the modular capabilities of AI agents.
Best use case
The primary use case for Skill Audit is to enhance the security posture of AI agent environments by proactively identifying and mitigating potential threats within third-party skills. It is most beneficial for AI agent developers, users, and platform maintainers who need to vet new or existing skills for malicious code, prevent data breaches, and ensure the integrity of their automated workflows.
ๆซๆ OpenClaw skills ไธญ็ๅฎๅ จ้ฃ้ฉ๏ผ้ฒๆญขไพๅบ้พๆปๅปใ
Users should expect a detailed security report outlining detected risks, their severity, and specific findings for the audited skill(s), along with a clear recommendation (e.g., 'DO NOT INSTALL').
Practical example
Example input
/skill-audit check suspicious-weather
Example output
๐ Skill Audit Report: suspicious-weather โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ Risk Score: 85/100 ๐ด HIGH RISK โโโโโโโโโโโโโโโฌโโโโโโโโโโโฌโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ File โ Severity โ Finding โ โโโโโโโโโโโโโโโผโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค โ index.ts โ CRITICAL โ Reads ~/.openclaw/credentials/ โ โ index.ts โ CRITICAL โ POST to webhook.site โ โ utils.ts โ WARNING โ Uses eval() โ โโโโโโโโโโโโโโโดโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ โ ๏ธ DO NOT INSTALL - This skill may steal your credentials!
When to use this skill
- Before installing a new AI agent skill from an unknown or untrusted source.
- To periodically audit all installed skills for latent vulnerabilities or updated malicious code.
- When developing a new skill, to self-check against common security pitfalls and best practices.
- After a security incident or suspected compromise, to investigate potentially affected skills.
When not to use this skill
- When the skill is confirmed to be running in a fully isolated and sandboxed environment.
- If the skill's source is already absolutely trusted and manually verified by security experts.
- For skills that do not involve code execution, file system access, or network requests (though this skill specifically targets code-based skills).
- If an audit is blocking a time-critical operation and the risk is deemed acceptable for a temporary, low-impact task.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/raini-skill-audit/SKILL.mdinside your project - Restart your AI agent โ it will auto-discover the skill
How Skill Audit ๐ Compares
| Feature / Agent | Skill Audit ๐ | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | easy | N/A |
Frequently Asked Questions
What does this skill do?
ๆซๆ OpenClaw skills ไธญ็ๅฎๅ จ้ฃ้ฉ๏ผ้ฒๆญขไพๅบ้พๆปๅปใ
How difficult is it to install?
The installation complexity is rated as easy. You can find the installation instructions above.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
Related Guides
AI Agents for Coding
Browse AI agent skills for coding, debugging, testing, refactoring, code review, and developer workflows across Claude, Cursor, and Codex.
Best AI Skills for Claude
Explore the best AI skills for Claude and Claude Code across coding, research, workflow automation, documentation, and agent operations.
Cursor vs Codex for AI Workflows
Compare Cursor and Codex for AI coding workflows, repository assistance, debugging, refactoring, and reusable developer skills.
SKILL.md Source
# Skill Audit ๐
ๆซๆ OpenClaw skills ไธญ็ๅฎๅ
จ้ฃ้ฉ๏ผ้ฒๆญขไพๅบ้พๆปๅปใ
---
## ๆไปค
### `/skill-audit scan [skill-name]`
ๆซๆๅทฒๅฎ่ฃ
็ skill๏ผๆฃๆตๅฏ็ไปฃ็ ๆจกๅผใ
```bash
# ๆซๆๆๆๅทฒๅฎ่ฃ
skill
skill-audit scan
# ๆซๆๆๅฎ skill
skill-audit scan moltdash
# ๆซๆๆฌๅฐ็ฎๅฝ
skill-audit scan ./my-skill
```
### `/skill-audit check <clawhub-slug>`
ๅฎ่ฃ
ๅๆฃๆฅ ClawHub ไธ็ skillใ
```bash
skill-audit check some-skill
```
---
## ๆฃๆต่งๅ
### ๐ด ้ซ้ฃ้ฉ (Critical)
- ่ฏปๅๅญ่ฏๆไปถ: `~/.ssh/`, `~/.env`, `credentials.json`
- ๅคๅๆฐๆฎ: `fetch()`, `curl`, `webhook`, `POST` ๅฐๆช็ฅ URL
- ไปฃ็ ๆง่ก: `eval()`, `exec()`, `child_process`
- ่ฏปๅ็ฏๅขๅ้ไธญ็ๅฏ้ฅ: `process.env.API_KEY`
### ๐ ไธญ้ฃ้ฉ (Warning)
- ็ฝ็ป่ฏทๆฑๅฐ้็ฅๅๅๅ
- ๆไปถ็ณป็ป้ๅ: `fs.readdir()`, `glob`
- ๅจๆ require/import
- Base64 ็ผ็ ็ๅญ็ฌฆไธฒ (ๅฏ่ฝๆฏๆททๆท)
### ๐ก ไฝ้ฃ้ฉ (Info)
- ไฝฟ็จ shell ๅฝไปค
- ่ฏปๅ็จๆท็ฎๅฝๅค็ๆไปถ
- ๅคง้ไพ่ตๅ
---
## ่พๅบ็คบไพ
```
๐ Skill Audit Report: suspicious-weather
โโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
Risk Score: 85/100 ๐ด HIGH RISK
โโโโโโโโโโโโโโโฌโโโโโโโโโโโฌโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ File โ Severity โ Finding โ
โโโโโโโโโโโโโโโผโโโโโโโโโโโผโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโค
โ index.ts โ CRITICAL โ Reads ~/.openclaw/credentials/ โ
โ index.ts โ CRITICAL โ POST to webhook.site โ
โ utils.ts โ WARNING โ Uses eval() โ
โโโโโโโโโโโโโโโดโโโโโโโโโโโดโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโโ
โ ๏ธ DO NOT INSTALL - This skill may steal your credentials!
```
---
## ่ฟ่กๆนๅผ
่ฏฅ skill ้ๅธฆไธไธช CLI ่ๆฌ๏ผagent ๅฏ็ดๆฅ่ฐ็จ๏ผ
```bash
node {baseDir}/src/audit.js scan ~/.openclaw/workspace/skills/moltdash
node {baseDir}/src/audit.js scan --all
```
---
## ๅ่
- [OWASP LLM Top 10](https://owasp.org/www-project-top-10-for-large-language-model-applications/)
- [Moltbook Security Discussion](https://www.moltbook.com/post/cbd6474f-8478-4894-95f1-7b104a73bcd5)Related Skills
Compliance & Audit Readiness Engine
Your AI compliance officer. Guides startups and scale-ups through SOC 2, ISO 27001, GDPR, HIPAA, and PCI DSS โ from zero to audit-ready. No consultants needed.
Compliance Audit Generator
Run internal compliance audits against major frameworks without hiring a consultant.
AI Safety Audit
Comprehensive AI safety and alignment audit framework for businesses deploying AI agents. Built around the UK AI Security Institute Alignment Project standards (2026), EU AI Act requirements, and NIST AI RMF.
SX-security-audit
ๅ จๆนไฝๅฎๅ จๅฎก่ฎกๆ่ฝใๆฃๆฅๆไปถๆ้ใ็ฏๅขๅ้ใไพ่ตๆผๆดใ้ ็ฝฎๆไปถใ็ฝ็ป็ซฏๅฃใGit ๅฎๅ จใShell ๅฎๅ จใmacOS ๅฎๅ จใๅฏ้ฅๆฃๆต็ญใๆฏๆ CLI ๅๆฐใJSON ่พๅบใ้ ็ฝฎๆไปถใๅฝ็จๆท่ฆๆฑ"ๅฎๅ จๆฃๆฅ"ใ"ๆผๆดๆซๆ"ใ"ๆ้ๆฃๆฅ"ใ"ๅฎๅ จๅฎก่ฎก"ๆถไฝฟ็จๆญคๆ่ฝใ
HIPAA Compliance for AI Agents
Generate HIPAA compliance checklists, risk assessments, and audit frameworks for healthcare organizations deploying AI agents.
Data Governance Framework
Assess, score, and remediate your organization's data governance posture across 6 domains.
Cybersecurity Risk Assessment
You are a cybersecurity risk assessment specialist. When the user needs a security audit, threat assessment, or compliance review, follow this framework.
afrexai-cybersecurity-engine
Complete cybersecurity assessment, threat modeling, and hardening system. Use when conducting security audits, threat modeling, penetration testing, incident response, or building security programs from scratch. Works with any stack โ zero external dependencies.
clickhouse-github-forensics
Query GitHub event data via ClickHouse for supply chain investigations, actor profiling, and anomaly detection. Use when investigating GitHub-based attacks, tracking repository activity, analyzing actor behavior patterns, detecting tag/release tampering, or reconstructing incident timelines from public GitHub data. Triggers on GitHub supply chain attacks, repo compromise investigations, actor attribution, tag poisoning, or "query github events".
security-guardian
Automated security auditing for OpenClaw projects. Scans for hardcoded secrets (API keys, tokens) and container vulnerabilities (CVEs) using Trivy. Provides structured reports to help maintain a clean and secure codebase.
mema-vault
Secure credential manager using AES-256 (Fernet) encryption. Stores, retrieves, and rotates secrets using a mandatory Master Key. Use for managing API keys, database credentials, and other sensitive tokens.
guardian-wall
Mitigate prompt injection attacks, especially indirect ones from external web content or files. Use this skill when processing untrusted text from the internet, user-uploaded files, or any external source to sanitize content and detect malicious instructions (e.g., "ignore previous instructions", "system override").