claude-code-skills
Warn
Audited by Socket on Oct 1, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: The skill's purpose is coherent, but it normalizes installing arbitrary third-party skills by cloning unpinned GitHub repos into Claude's auto-load directories and points to a personal repo for expanded resources. No direct credential theft or malicious execution is present, but the transitive trust and supply-chain exposure are disproportionate enough to warrant caution.
Confidence: 91%Severity: 74%
Audit Metadata