ClawHub
Back to skill

Security audit

Openclaw Safety Coach

Security checks across malware telemetry and agentic risk

Overview

This skill is a security coaching guide that gives defensive OpenClaw safety instructions and does not contain executable code or hidden data access.

Install this if you want stricter OpenClaw safety coaching and refusal behavior. Review suggested commands before running them, especially configuration-changing commands such as `openclaw security audit --fix`, because the skill itself does not execute them but it may recommend them.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (1)

Vague Triggers

Medium
Confidence
89% confidence
Finding
The README states that the Safety Coach 'automatically activates when potentially unsafe requests are detected' without defining precise activation criteria, priority, or boundaries. In a safety-focused skill, ambiguous triggering can cause overbroad interception, inconsistent enforcement, or failure to engage when needed, which can be abused by adversarial prompt phrasing or lead operators to overtrust protections that are not deterministic.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.