Video Assembly
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process external content such as caption sources and media URLs, creating a surface for indirect instructions to be embedded in data.\n
- Ingestion points: The skill accepts clip URLs and caption source text as inputs (SKILL.md).\n
- Boundary markers: The instructions lack specific delimiters or boundary markers to isolate external data from the agent's core logic.\n
- Capability inventory: The skill uses media.combine_video and media.upload_file capabilities (SKILL.md).\n
- Sanitization: The skill includes steps to proofread transcription timing and names but does not mention filtering for malicious instructions (SKILL.md).
Audit Metadata