director

Pass

Audited by Gen Agent Trust Hub on Apr 12, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted user data including scene prompts and character JSON files. Ingestion points: User-provided character.json and scene prompts. Boundary markers: No delimiters or instructions to ignore embedded commands are present in the orchestration logic. Capability inventory: The pipeline involves file processing and uploading via the editor-post-production skill. Sanitization: Input validation or sanitization of external content is not explicitly performed.
  • [EXTERNAL_DOWNLOADS]: The skill references VidJutsu documentation for API integration.
  • [SAFE]: Required API keys are managed through standard environment variable configuration rather than hardcoding.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 12, 2026, 01:54 AM