drawing-video
Warn
Audited by Socket on Jul 30, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
SUSPICIOUS. The media-processing purpose is coherent, and local ffmpeg/PIL usage is proportionate, but the skill’s Gemini integration is routed through a custom Cloudflare gateway with auto-read proxy credentials instead of direct official Google API calls. That intermediary data flow, plus reliance on sibling skills, creates a medium-high security risk without enough evidence for confirmed malware.
Confidence: 87%Severity: 74%
Audit Metadata