brand-guidelines

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOW
Full Analysis
  • NO_CODE (SAFE): The skill contains only Markdown documentation and YAML metadata. There are no associated scripts, binaries, or executable commands provided in the analyzed file.
  • EXTERNAL_DOWNLOADS (SAFE): While the documentation mentions 'python-pptx' and specific fonts, it does not include instructions to download packages, curl scripts, or fetch remote resources at runtime.
  • DATA_EXFILTRATION (SAFE): There are no file access or network operations defined that could lead to data exposure or exfiltration.
  • PROMPT_INJECTION (SAFE): The instructions are descriptive of styling rules (colors, fonts) and do not contain patterns aimed at bypassing agent safety filters or overriding system prompts.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 12:59 AM