Back to skill
Skillv0.1.0
VirusTotal security
Agent Security Patterns · External malware reputation and Code Insight signals for this exact artifact hash.
Scanner verdict
BenignApr 30, 2026, 4:58 AM
- Hash
- 71b90512a395c7deff574db48b7f00c9210810ffa6bb7987cef3c38a9bdf8f2d
- Source
- palm
- Verdict
- benign
- Code Insight
- Type: OpenClaw Skill Name: security-awareness-skill Version: 0.1.0 This skill is designed to educate an AI agent on recognizing and responding to potentially malicious skill patterns, including prompt injection, obfuscated payloads, external fetches, and credential requests. All examples of malicious patterns are presented within an educational context, clearly labeled as 'What it looks like:' and followed by 'Why it matters:' and 'Your reasoning check:'. The instructions given to the agent are defensive, guiding it to pause, describe concerns, ask for verification, and suggest alternatives when encountering suspicious patterns. There is no evidence of intentional harmful behavior, data exfiltration, or unauthorized execution within the skill itself; rather, it aims to prevent such actions by other skills. The `_meta.json` file contains standard metadata.
- External report
- View on VirusTotal