Description-Behavior Mismatch
Medium
- Confidence
- 91% confidence
- Finding
- The skill is framed as editing user-uploaded video, but the documented supported formats and upload behavior expand into arbitrary media types and URL-based ingestion. This broadens the trust boundary and can enable server-side fetching of attacker-controlled URLs, unexpected processing of non-video content, or user confusion about what data the skill may transmit to the backend.
