ClawGuard | Anti-Malware Scanner & Honeypot

Security checks across malware telemetry and agentic risk

Overview

ClawGuard is a disclosed local security-monitoring skill that creates a decoy honeypot file, scans OpenClaw skills, and only uses shell actions after approval.

Install only if you want recurring local monitoring of OpenClaw skills. Expect a fake credential-like honeypot file at ~/.openclaw/workspace/routing_config_backup.json, and review any alert carefully before approving quarantine actions.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
83% confidence
Finding
The skill runs on a cron schedule every 10 minutes and scans newly added directories, but the trigger scope and file selection criteria are only loosely described. Scheduled security-monitoring skills with broad recurring execution can create continuous surveillance behavior and unexpected access patterns, especially when combined with read/write and notification capabilities.

Missing User Warnings

Medium
Confidence
92% confidence
Finding
The skill creates a honeypot file populated with dummy internal tokens and webhook endpoints, but this deceptive artifact is not disclosed in the user-facing description. Even if the contents are fake, silently planting credential-like material can mislead users, trigger other tools, or be abused as cover for harvesting or exfiltration logic in a supposedly defensive skill.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal