The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external media assets (video and 3D files) which constitutes a potential attack surface for instructions embedded in file metadata or content.
Ingestion points: Media assets are brought into the environment via upload_asset and accessed using asset_get as described in SKILL.md.
Boundary markers: The instructions do not define specific delimiters or guardrails to prevent the agent from interpreting metadata or embedded content as instructions.
Capability inventory: The skill utilizes capabilities including model_run for execution, asset_download for file retrieval, and asset_display for previewing results.
Sanitization: No specific sanitization or validation steps for the content of the uploaded media files are mentioned.

scenario-video-to-motion