avatar-single-scene
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a comprehensive intake and approval process to gate the generation of AI assets.
- [EXTERNAL_DOWNLOADS]: Downloads specialized sub-skills from the vendor's official GitHub repository using npx.
- [COMMAND_EXECUTION]: Instructions involve using curl for interacting with the pruna-api and ffprobe for audio inspection.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided dialogue scripts and visual prompts. 1. Ingestion points: User-supplied voice_script, voice_prompt, and video_prompt. 2. Boundary markers: Mandatory approval gates (approve plan, approve still, approve clips). 3. Capability inventory: Uses curl for file uploads and prediction requests to the vendor API. 4. Sanitization: No explicit text sanitization or filtering is mentioned in the instructions.
Audit Metadata