performance-scene-director
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized access attempts were found. The skill is entirely descriptive and instruction-based, focused on design patterns for acting and cinematography.
- [INDIRECT_PROMPT_INJECTION]: The skill defines an interface for processing user-supplied emotional scene descriptions and external reference materials.
- Ingestion points: Scene descriptions in the prompt and external materials referenced via the
@material[...]tag inSKILL.md. - Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded commands in the material references.
- Capability inventory: The skill has no tools or permissions for file system modification, network requests, or shell command execution.
- Sanitization: No input sanitization is present. However, because the skill only outputs text prompts for downstream AI models and lacks high-privilege capabilities, the attack surface is not exploitable in this context.
Audit Metadata