flag-publish
Warn
Audited by Socket on Jul 7, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS: the stated purpose aligns with GrowthBook flag publishing, and the API/domain targets look legitimate, but the skill’s core operation is routed through an unseen bundled helper that receives GB_API_KEY. That credential-forwarding to unverifiable local code makes the trust footprint disproportionate enough to rate as high security risk even without evidence of confirmed malware.
Confidence: 84%Severity: 82%
Audit Metadata