SkillGuard is a dedicated security scanner that audits Openclaw Skills for dangerous patterns, vulnerable dependencies, and suspicious behaviors before installation.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install skill-guard-security
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install skill-guard-security using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is SkillGuard?
SkillGuard acts as a critical security layer for the AI agent ecosystem, specifically designed to analyze Openclaw Skills from ClawHub or local directories. It performs static analysis to detect high-risk operations such as command injections, unsafe eval usage, and unauthorized network or file system access, ensuring that every skill you add to your agent is vetted for safety.
By providing a transparent risk assessment, SkillGuard helps developers and organizations maintain a secure environment when extending their AI agents. It bridges the gap between functionality and security, making it a cornerstone tool for anyone building or consuming Openclaw Skills.
SkillGuard Use Cases
- Scanning third-party agent skills from ClawHub before performing a production installation.
- Auditing local development directories to ensure custom Openclaw Skills adhere to security best practices.
- Automating security checks within a CI/CD pipeline for agent skill deployment.
- Running a centralized security API to validate skills across a team or organization.
How SkillGuard Works
- The user initiates a scan via the CLI or API by pointing to a specific skill slug or local folder.
- SkillGuard parses the source code to identify dangerous patterns like shell execution or network call vulnerabilities.
- The tool cross-references dependencies against known vulnerability databases to find outdated or compromised packages.
- A security verdict is calculated based on the severity and frequency of detected risks.
- The tool outputs a detailed JSON report or terminal summary, categorizing the skill as SAFE, CAUTION, or DANGEROUS.
SkillGuard Setup
To perform a quick audit of any Openclaw Skills, use the following commands:
# Audit a specific skill by its name
npx skillguard-audit --name <skill-slug>
# Audit a skill located in your current directory
npx skillguard-audit --path ./my-skill
# Start the security API server
npx skillguard-audit serve --port 3402
SkillGuard Data Schema & Taxonomy
SkillGuard provides a structured JSON output for every audit performed on Openclaw Skills. The schema is organized as follows:
| Property | Type | Description |
|---|---|---|
| skill | string | The identifier of the audited skill |
| verdict | string | The security rating: SAFE, CAUTION, or DANGEROUS |
| score | number | A risk score from 0 to 100 |
| risks | array | A list of objects containing risk type, severity, file path, and line number |
SkillGuard Advanced Features
- Multi-channel auditing via both CLI and a dedicated REST API server.
- Real-time detection of suspicious shell commands and file system access risks.
- Deep integration capabilities for marketplaces and hubs hosting Openclaw Skills.
- Precise risk location reporting including file paths and specific line numbers for rapid remediation.
SKILL.md
Loading
Related Openclaw Skills
ClkLog Local Deployment
ClkLog is a professional open-source user behavior analysis and profiling system utilizing ClickHouse for high-performance data storage.
Dawang Workspace
Dawang Workspace is a specialized execution environment for AI agents focused on code generation, technical implementation, and dietary data visualization.
Fortune Master Suanming
A professional-grade Bazi and Four Pillars of Destiny analysis tool that provides logical, evidence-based metaphysical insights.
UK Employee Handbook & Policy Generator
An automated tool for generating comprehensive, legally compliant UK employee handbooks tailored to business size and industry requirements.
AWS CLI
A repackaged AWS CLI v2 skill for OpenClaw agents, providing official Amazon Web Services command-line capabilities without requiring a system Python installation.
A-Stock Investment Analyst
A comprehensive A-share market analysis tool providing fundamental, technical, and capital flow insights for professional investors.
Featured*
