watching-videos
Warn
Audited by Socket on Aug 22, 2026
1 alert found:
SecuritySecuritySKILL.md
MEDIUMSecurityMEDIUM
SKILL.md
The stated purpose is coherent with the media-processing behavior, but the skill relies on an unverifiable external `watch-skill` binary and an auto-installer that can add dependencies without provenance details. That makes it suspicious from an install-trust perspective, with additional medium risk from processing untrusted multimedia content into agent-readable text.
Confidence: 86%Severity: 78%
Audit Metadata