brand-guidelines
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [Prompt Injection] (SAFE): The instructions are strictly limited to brand styling and contain no override or bypass markers.
- [Data Exposure & Exfiltration] (SAFE): No sensitive file paths, credentials, or network operations are present.
- [Remote Code Execution] (SAFE): No scripts or remote execution patterns are included; the skill is descriptive markdown.
- [Obfuscation] (SAFE): No encoded content or hidden characters were detected.
- [Indirect Prompt Injection] (SAFE): While the skill describes how to style content, it does not ingest or process untrusted data directly in its current state.
Audit Metadata