feature-design
Warn
Audited by Socket on Mar 29, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the visible workflow is coherent with feature design, and there are no explicit network calls or credential grabs, but the skill's core behavior is delegated to undisclosed ~/.claude/scripts helpers. That unverifiable local-code dependency makes install/execution trust weak and raises overall risk despite otherwise proportionate purpose and local-only data flow.
Confidence: 84%Severity: 72%
Audit Metadata