An advanced security management system for OpenClaw extensions that provides automated risk scanning and granular policy control.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install skill-shield-007
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-shield-007 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 Skill Shield?
Skill Shield is a dedicated security infrastructure designed to monitor and manage the safety of installed extensions within the OpenClaw ecosystem. It acts as a protective layer that analyzes code for potential vulnerabilities, ensuring that every interaction with Openclaw Skills remains secure and transparent.
By implementing a sophisticated risk assessment engine, Skill Shield identifies critical threats such as unauthorized network access, command execution, and credential theft. It empowers developers to maintain a secure workspace through customizable allowlists and automated prompts for high-risk operations.
Skill Shield Use Cases
- Automatically scanning all installed Openclaw Skills for security vulnerabilities upon installation.
- Enforcing mandatory user confirmation before executing extensions with high-risk permissions like network access or file writing.
- Creating a trusted environment by managing a global allowlist for frequently used, verified extensions.
- Auditing historical security decisions and blocked execution attempts for compliance and safety monitoring.
How Skill Shield Works
- The system traverses the local skills directory to identify every installed extension and its associated metadata.
- It performs a keyword-based code analysis on scripts and reference files to detect 10 distinct categories of security risks.
- A severity score is calculated (Severe, High, Medium, Low) based on the specific threats found within the extension.
- Before any extension runs, the shield checks against the allowlist and blacklist configurations.
- If a risk is identified, the system prompts the user for manual approval or blocks execution entirely depending on the severity level.
- All scan results and user decisions are persisted in a centralized memory database for future reference.
Skill Shield Setup
To set up Skill Shield, ensure you have Python 3.6+ installed. You can configure the system by editing the global config file or using CLI commands.
# Check the extension directory permissions
ls -la ~/.openclaw/workspace/skills/
# Manually trigger a security scan of all extensions
python3 ~/.openclaw/workspace/skills/skill-shield/scripts/shield.py scan
Skill Shield Data Schema & Taxonomy
Skill Shield organizes its data using a structured JSON schema stored in the workspace memory. This includes risk reports and historical audit logs.
Risk Taxonomy
| Risk ID | Risk Name | Severity | Keywords Detected |
|---|---|---|---|
| R001 | Network Access | High | requests, curl, fetch |
| R004 | Command Execution | Severe | exec, subprocess, system |
| R007 | Credential Access | Severe | apiKey, token, secret |
Configuration Schema
- allowlist: Array of trusted extension names that bypass checks.
- blocked: Array of blacklisted extensions that are strictly prohibited.
- scanOnInstall: Boolean flag to enable automatic scanning during installation.
Skill Shield Advanced Features
- Granular Severity Mapping: Custom triggers for mandatory user confirmation based on risk levels (R001-R010).
- Persistence Layer: Long-term storage of risk reports and decision history in
shield-risks.json. - Blacklist Enforcement: Hard-blocking of dangerous extensions to prevent any accidental execution.
- Automated Workflow Hooks: Integration plans for automatic scanning during clauhub installation and periodic security reporting.
SKILL.md
Loading
Related Openclaw Skills
3-Layer Token Compressor
A local preprocessing utility that leverages Ollama models to compress prompts and summarize history before they reach expensive paid LLM APIs.
PinchTab Browser Control
A lightweight and token-efficient browser automation tool that allows AI agents to control web pages using stable accessibility references.
Docupilot
A document automation integration for generating and managing documents from templates within AI agent workflows.
Cardinal
A powerful project management skill that allows AI agents to organize tasks, track progress, and manage resources within Cardinal.
ClawHub Publisher
ClawHub Publisher is a professional developer tool that automates the entire lifecycle of publishing and managing Openclaw Skills.
Doubao Seed Image Analysis Skill
A specialized CLI tool that leverages ByteDance Doubao vision models to analyze and describe image content for users and AI agents.
METADATA
Featured*
