sound-designer
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized data access were detected. The instructions are focused on audio engineering parameters such as loudness targets (-14 LUFS), EQ settings, and mixing techniques (ducking, layering).
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes external content.
- Ingestion points: Reads timeline clip data and source levels via
audio_inspect(SKILL.md) and processes user-provided project briefs. - Boundary markers: No specific boundary markers or "ignore instructions" warnings are defined for processed data.
- Capability inventory: Includes tools for fetching stock media, generating media, and performing various audio manipulations (
autoDuck,setAudio,detachAudio). - Sanitization: No explicit sanitization or validation of the input data or metadata is mentioned.
Audit Metadata