vos-migrate
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external media files (e.g., demo.mp4), which creates an ingestion point for potentially untrusted data that could attempt to influence the agent's behavior through crafted metadata or visual content.
- Ingestion points: External video files processed via
ffprobeandffmpeginSKILL.md. - Boundary markers: Absent; the skill does not define specific delimiters for the data extracted from media files.
- Capability inventory: File system modification (creating directories and writing media files) and network operations (pushing data via the
vosCLI). - Sanitization: Absent; the workflow relies on the agent's direct interpretation of tool outputs and generated stills.
Audit Metadata