Motion Control
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection through the processing of untrusted external media.
- Ingestion points: The skill processes user-provided movement references, driving videos, and frame images as specified in the Required inputs section of SKILL.md.
- Boundary markers: There are no explicit boundary markers or instructions provided to the agent to ignore embedded commands within the visual or metadata context of the processed assets.
- Capability inventory: The skill utilizes media.upload_file, media.generate_video, and media.check_result to handle and process these external assets.
- Sanitization: Instructions focus on quality assurance and likeness review rather than programmatic sanitization or validation of input media to detect adversarial patterns.
Audit Metadata