typesetting

Pass

Audited by Gen Agent Trust Hub on Sep 10, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and directs the agent to fetch documentation and guidelines from well-known services and trusted organizations, including:
  • W3C technical reports (w3.org)
  • Digital Agency of Japan (design.digital.go.jp)
  • UK Government Design System (service.gov.uk)
  • SmartHR Design System (smarthr.design)
  • [INDIRECT_PROMPT_INJECTION]: The typesetting workflow instructs the agent to inspect real references from external sites (Category 8c). This ingestion of external, uncontrolled content for the purpose of extracting 'composition rules' or 'specifications' creates a surface where a compromised third-party page could attempt to influence the agent's behavior through instructions embedded in the documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 10, 2026, 01:24 PM