verify-bug
Warn
Audited by Socket on May 15, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The skill is broadly aligned with UAT verification and uses official vendor ecosystems, so it does not look malicious. However, it asks the agent to extract credentials from Confluence and to autonomously modify JIRA workflow state, which is a disproportionate and sensitive operational footprint for a verification skill.
Confidence: 87%Severity: 58%
Audit Metadata