Back to skill
Skillv1.0.0
VirusTotal security
Agent Audit · External malware reputation and Code Insight signals for this exact artifact hash.
Scanner verdict
BenignApr 30, 2026, 4:09 AM
- Hash
- ae73ef41e670bf8a1607ae513305ebc125875337299a3b1b5f151f0f9a4ec6ef
- Source
- palm
- Verdict
- benign
- Code Insight
- Type: OpenClaw Skill Name: agent-audit Version: 1.0.0 The OpenClaw Agent Audit skill is designed to analyze agent configurations and usage for cost optimization. It reads OpenClaw configuration files (e.g., `~/.openclaw/openclaw.json`) to identify models and agent setups, which is necessary for its stated purpose. The `scripts/audit.py` code does not perform any network requests, execute arbitrary system commands, or modify any files other than optionally writing a markdown report to a user-specified path. The `SKILL.md` instructions are clear, align with the script's functionality, and contain no evidence of prompt injection attempts to subvert the agent's behavior for malicious purposes. The skill explicitly states it is 'read-only' and the code confirms this, focusing solely on analysis and reporting without making changes or exfiltrating sensitive data.
- External report
- View on VirusTotal