githuman
Warn
Audited by Socket on Sep 24, 2026
1 alert found:
AnomalyAnomalySKILL.md
LOWAnomalyLOW
SKILL.md
SUSPICIOUS. The skill’s purpose and local data flows are coherent for a code-review helper, and there is no sign of credential theft or off-platform exfiltration. However, install trust is weakened by publisher/package ambiguity around `githuman` vs `@tequila99/githuman` and by unpinned npm execution, so the skill is best classified as medium-risk suspicious rather than benign.
Confidence: 87%Severity: 52%
Audit Metadata