sonic-branding
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze external brand-related documents, which could potentially contain malicious instructions.
- Ingestion points: Step 1 (Brand Assessment) requires reviewing external brand guidelines, mission statements, and target audience research. Step 4 (Sound Mood Board) involves processing audio links and references.
- Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are defined for the ingested brand data.
- Capability inventory: The skill is primarily focused on text and strategy generation (briefs, templates, audit summaries). It does not exhibit dangerous capabilities like shell command execution, file system writes, or network requests based on the ingested content.
- Sanitization: No explicit sanitization or validation steps are defined for the brand inputs before they are interpolated into the sonic strategy framework.
Audit Metadata