A comprehensive security suite for Openclaw Skills that provides real-time threat detection, malware scanning, and agent integrity monitoring.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install crusty-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 crusty-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 Crusty Security?
Crusty Security is a specialized defense layer designed specifically for Openclaw Skills users. It addresses the unique threat model of AI agents, where prompt injections and malicious third-party skills can lead to system compromise or unauthorized data exfiltration. By integrating ClamAV for robust file scanning and custom static analysis for skill auditing, it ensures that both the host environment and the agent's internal logic remain untampered.
This skill provides a vital security framework for Openclaw Skills, protecting the agent's integrity by monitoring core files like SOUL.md and MEMORY.md for suspicious churn. Whether you are running an agent on a high-end server or a resource-constrained device, Crusty Security provides the necessary tools to maintain a hardened security posture.
Crusty Security Use Cases
- Scan downloaded files and directories for malware and viruses automatically.
- Audit third-party skills from ClawHub before installation to prevent supply chain attacks.
- Monitor agent configuration files for unauthorized modifications or prompt injection indicators.
- Perform deep host security audits to identify suspicious cron jobs, listening ports, or SSH keys.
- Generate detailed security posture reports to track the health of your Openclaw Skills environment.
How Crusty Security Works
- The security workflow is initiated via specific triggers or scheduled cron jobs managed within the Openclaw Skills ecosystem.
- The system utilizes ClamAV signatures to perform deep file inspections for known malware and suspicious binary patterns.
- Static analysis scripts evaluate skill directories for high-risk behaviors such as reverse shells, crypto-mining, or credential harvesting.
- Host-level monitoring checks system integrity, looking for unauthorized outbound connections or hidden persistence mechanisms.
- If threats are detected, the skill provides a quarantine workflow, moving dangerous files to a secure location while maintaining a metadata manifest.
- Results are compiled into JSON logs and can be optionally pushed to a centralized dashboard for real-time monitoring of all Openclaw Skills instances.
Crusty Security Setup
To begin securing your Openclaw Skills environment, execute the main setup script which handles dependency installation and initial configuration.
# Navigate to the skill directory and run setup
bash setup.sh
After installation, you must configure the necessary cron jobs using the agent's internal tools to ensure continuous protection. You can verify the status of your security tasks with the following command:
bash scripts/check_crons.sh
Ensure that environment variables like CRUSTY_API_KEY are set if you intend to use the cloud dashboard features.
Crusty Security Data Schema & Taxonomy
The skill maintains a structured data organizational system to track threats and audits within Openclaw Skills.
| Component | Path | Description |
|---|---|---|
| Quarantine | /tmp/crusty_quarantine |
Secure storage for malicious or suspicious files. |
| Manifest | manifest.json |
JSON database tracking original paths and timestamps of quarantined items. |
| Logs | /tmp/crusty_logs |
Directory containing historical scan results and security reports. |
| Configuration | Environment Variables | Uses CRUSTY_ prefixed variables for workspace paths and API keys. |
Crusty Security Advanced Features
- Incremental scanning mode that significantly reduces processing time by only inspecting changed files within your Openclaw Skills workspace.
- Resource-constrained optimization for Raspberry Pi, allowing security monitoring on devices with less than 1GB of RAM.
- Automated skill auditing that detects complex obfuscation techniques and unauthorized system file access in new integrations.
- Dashboard integration for multi-agent environments, providing a single pane of glass for all your Openclaw Skills security events.
- Quarantine management system with manifest tracking to prevent accidental data loss during remediation.
SKILL.md
Loading
Related Openclaw Skills
Capture Website
A command-line utility for capturing web screenshots and sharing them through automated messaging workflows.
Brawlnet
An official tactical combat protocol designed for autonomous agents to compete in the fast-paced BRAWLNET hex-grid arena.
moltbook
A specialized skill enabling AI agents to interact with, strategize for, and commentate on live MolterStrike CS 1.6 bot matches.
DECK-0
Enable AI agents to interact with the DECK-0 digital collectibles platform to manage, buy, and open trading card packs on-chain.
Clawswitch
A natural language model router designed to swap AI backends instantly within the OpenClaw environment without manual configuration.
Silk: Agent Banking & Payments on Solana
Silk is a non-custodial banking and payment layer for AI agents on the Solana blockchain, enabling secure USDC transfers and automated escrows.
METADATA
Featured*
