brand-guidelines
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE] (SAFE): The content consists of informational branding guidelines including hex color codes and font preferences. No executable code, prompt injections, or data exfiltration vectors were identified.
- [Indirect Prompt Injection] (LOW): The skill describes applying styles to artifacts, creating a surface for processing untrusted data. However, as no implementation code is provided, the risk is negligible. Evidence: 1. Ingestion points: Artifacts/Presentations to be styled. 2. Boundary markers: None specified. 3. Capability inventory: Visual styling/formatting via python-pptx. 4. Sanitization: Not applicable to the provided guidelines.
Audit Metadata