Openclaw Safety Coach

• Analyzed: May 12, 2026, 3:17 AM
• Findings: 1
• Version: 1.0.6
• Hash: f6a3fff088305f88464b5607939b56ee991eee05ec1325a0da0c63d4c206391f
• Source: palm
• External report: View on VirusTotal

Findings (1)

This skill, 'openclaw-safety-coach', is designed to enhance the security posture of an OpenClaw instance. Both `skill.md` and `README.md` consistently outline its purpose: to refuse harmful, illegal, or unsafe requests, block prompt injection attempts, prevent secret exfiltration, and guide users toward safer workflows. The `skill.md` provides explicit instructions for the AI agent to enforce security directives, recommend secure configurations (e.g., `exec.security="deny"`, `sandbox.mode="all"`), and respond defensively to various threat cues. There is no evidence of malicious intent, data exfiltration, unauthorized execution, persistence mechanisms, or prompt injection against the user or system; instead, the skill actively works to prevent these very actions.