build-skills
Warn
Audited by Socket on Mar 13, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the stated purpose is coherent, but the skill materially expands trust boundaries by pulling in untrusted remote skill content and using it to shape local file-writing decisions. The biggest issue is indirect prompt-injection exposure during remote research, plus moderate supply-chain risk from the unspecified skill-dl/tooling path.
Confidence: 85%Severity: 76%
Audit Metadata