maintaining-brand-identity
Pass
Audited by Gen Agent Trust Hub on Feb 16, 2026
Risk Level: LOW
Full Analysis
- [DATA_EXPOSURE & EXFILTRATION] (SAFE): No sensitive file paths, hardcoded credentials, or network operations were detected. The skill only contains public-facing design metadata.
- [PROMPT_INJECTION] (SAFE): The instructions use standard instructional language to enforce brand consistency. No patterns indicative of system prompt overrides or safety filter bypasses were found.
- [REMOTE_CODE_EXECUTION] (SAFE): The skill does not perform any downloads or remote script execution. It references standard libraries (React, Tailwind) as architectural constraints for the agent's output rather than performing runtime installation.
- [INDIRECT_PROMPT_INJECTION] (LOW): The skill processes local resource files to inform agent reasoning. While these files define the agent's persona and output style, they do not ingest untrusted external data and the skill lacks the capabilities (file writes, execution, network) to be exploited via this vector.
Audit Metadata