director-master
Pass
Audited by Gen Agent Trust Hub on Sep 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided screenplays, outlines, and summaries as primary inputs to generate storyboards, which allows potential instructions embedded in those scripts to influence agent behavior. + Ingestion points: User-supplied script text and story outlines are ingested at the beginning of the workflow as specified in SKILL.md. + Boundary markers: The prompt instructions do not utilize specific delimiters or isolation techniques to prevent the agent from interpreting instructions contained within user content. + Capability inventory: The skill performs stylistic analysis and structured data generation (Markdown and XLSX); it does not have access to network operations or arbitrary code execution environments. + Sanitization: No filtering or validation mechanisms are defined to detect or strip prompt injection attempts from the input text.
- [EXTERNAL_DOWNLOADS]: The README.md file contains a reference link to a related screenwriting project hosted on GitHub, a well-known service.
Audit Metadata