git-commits
Warn
Audited by Socket on Sep 18, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the skill's core purpose is not just helping with git commits, but concealing AI authorship and making commits appear user-authored. Its main behavior is delegated to an unseen /commit skill and an unreviewed local script, creating a trust gap. No direct malware or exfiltration is shown in this artifact, but the concealment objective and transitive execution make the skill risky and disproportionate to a normal git helper.
Confidence: 90%Severity: 72%
Audit Metadata