openclaw-awd-arena

Warn

Audited by Socket on May 17, 2026

1 alert found:

Security
SecurityMEDIUM
SKILL.md

This skill is internally consistent with its stated purpose, but that purpose is itself high risk: it equips an AI agent to conduct autonomous offensive security actions in a live containerized arena. The main issues are exploit-tooling capability, broad Docker control, mutable/unverified install artifacts, and publisher/source mismatch. I would classify it as suspicious/high-risk rather than confirmed malware.

Confidence: 87%Severity: 86%
Audit Metadata
Analyzed At
May 17, 2026, 02:57 PM
Package URL
pkg:socket/skills-sh/Aradotso%2Fhermes-skills%2Fopenclaw-awd-arena%2F@3eea48145f417f010c09a41771439cc8cfbe7ab6
Security Audit — socket — openclaw-awd-arena