qa-agent-claude
Warn
Audited by Socket on Jul 8, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS: The skill is broadly aligned with QA automation and uses official Playwright tooling, so there is no strong malware or credential-harvesting signal. However, its autonomous closed-loop behavior, ability to drive real applications, mutate repo files, and process untrusted app content make it a medium-risk skill that exceeds a simple documentation/reference role.
Confidence: 85%Severity: 56%
Audit Metadata