The Agent Skills Directory

[SAFE]: No security issues were identified in the skill. The instructions strictly follow the intended purpose of managing video clips through the Montage MCP server.
[SAFE]: Network activity is restricted to the legitimate service domain studio.montage.app, which is necessary for the skill's core functionality.
[SAFE]: The skill processes external data (transcripts and video metadata), which is a standard requirement for its operation. While this represents a surface for potential indirect prompt injection, the skill does not possess high-risk capabilities like arbitrary shell execution or file system access that would make such an injection exploitable.
Ingestion points: get_transcript, get_video_corpus, and get_chapters in SKILL.md retrieve external content.
Boundary markers: Not explicitly defined.
Capability inventory: Limited to API tool calls for video metadata retrieval and clip creation (create_moment).
Sanitization: Not explicitly specified in the skill body.

create-moment