Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
What This Skill Does
Scans installed OpenClaw skills for suspicious code patterns, cross-references file hashes with VirusTotal, and resolves false positives by adding verified files to the security allowlist. Malicious files are moved to a quarantine folder rather than deleted.
Combines local heuristic scanning with live VirusTotal reputation checks in one command, avoiding the need to manually hash files and query the API separately.
When to use it
- Auditing newly installed third-party skills before trusting them
- Clearing false-positive security flags on known-safe skills
- Quarantining skills flagged as malicious by VirusTotal
- Running a dry-run security check before committing changes to the allowlist
- Automating periodic trust verification across a skills directory
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: verify my installed skills and allowlist anything that's clean
- 1Run heuristic scan on the skills/ directory to identify suspicious patterns
- 2Compute hashes for flagged files
- 3Cross-reference each hash against VirusTotal for reputation data
- 4Allowlist files with zero detections via the Gateway RPC Bridge
- 5Move files with detections greater than zero to .quarantine/
Suspicious skills are either allowlisted as safe or quarantined, with a report of actions taken
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.