Back to skill
Skillv1.0.32
VirusTotal security
Agentshield Audit · External malware reputation and Code Insight signals for this exact artifact hash.
Scanner verdict
BenignApr 29, 2026, 4:16 AM
- Hash
- 044933edd0f2904efa628e375aa70d4610cd7df63ecd35e941f2d5dc998448b0
- Source
- palm
- Verdict
- benign
- Code Insight
- Type: OpenClaw Skill Name: agentshield-audit Version: 1.0.32 This bundle provides a comprehensive security auditing and cryptographic identity framework for AI agents, including modules for secret scanning, input sanitization, and a 'Trust Handshake' protocol. While the tool performs sensitive operations such as scanning the local workspace for API keys (`secret_scanner.py`) and communicating with an external API (`agentshield.live`), the implementation includes explicit data sanitization whitelists in `audit_client.py` to prevent the exfiltration of sensitive payloads, agent responses, or evidence. The logic is consistent with its stated purpose of hardening agent security and establishing verifiable identities, and the code includes human-in-the-loop consent prompts for file access.
- External report
- View on VirusTotal