brand-guidelines-enforcer

Pass

Audited by Gen Agent Trust Hub on Feb 26, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect prompt injection surface identified. The skill is designed to ingest and analyze content from a local directory, which may contain untrusted data or adversarial instructions designed to influence the agent's output.
  • Ingestion points: Source code files and UI labels located in the src/ directory referenced in the SKILL.md usage example.
  • Boundary markers: Absent; the documentation provides no evidence of delimiters or specific instructions for the agent to disregard instructions found within the analyzed files.
  • Capability inventory: The skill is capable of reviewing copy tone, validating compliance, and suggesting changes to UI text.
  • Sanitization: No content sanitization or validation mechanisms are described for the input data.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 26, 2026, 08:36 AM