brand-voice
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill focuses on linguistic analysis and content generation guidelines. It includes a specific security and privacy measure in the 'Persistence Rules' section, instructing the agent not to store personal voice fingerprints in shared repository-tracked files without the user's explicit request.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external text samples which represents a potential injection surface.
- Ingestion points: The skill ingests user-provided text such as social media posts, emails, and articles as defined in the 'Source Priority' section of SKILL.md.
- Boundary markers: The instructions lack explicit markers or warnings to disregard potential instructions embedded within the source materials.
- Capability inventory: The skill's capabilities are limited to text analysis and extraction; it does not utilize shell execution, file system modifications outside of the workspace, or unauthorized network calls.
- Sanitization: There is no sanitization of the source content, but the risk is mitigated by the structured nature of the extraction task, which focuses on stylistic metrics rather than command interpretation.
Audit Metadata