director-agent

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied scripts and story outlines as its primary input, creating a surface for potential indirect prompt injection. This data is used to generate director analyses and influence web research queries.- [EXTERNAL_DOWNLOADS]: The skill includes references to external resources for film theory and workflow inspiration.
  • References deep-knowledge cards from a companion GitHub repository.
  • References several well-known GitHub projects (e.g., Storyboarder, VideoClaw, ViMax) as models for production workflows.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 10:13 AM
Security Audit — agent-trust-hub — director-agent