shatter
Warn
Audited by Socket on Apr 27, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill is internally coherent as a chaos-engineering/red-team-style resilience testing skill, with no evident credential theft or exfiltration path, but it grants an AI agent high-risk disruptive security-testing behaviors against real systems. The main concern is offensive/destructive capability, not supply-chain or data-flow abuse.
Confidence: 90%Severity: 74%
Audit Metadata