design-qa

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute several local Python and Node.js scripts (scripts/validate_tokens.py, scripts/validate_contrast.py, scripts/lint_hardcodes.py, scripts/measure_render.mjs). These scripts are intended for automated linting and rendering tasks within the project repository.
  • [INDIRECT_PROMPT_INJECTION]: The skill involves processing external data such as workflows/design-qa.md and user-specified source or HTML files.
  • Ingestion points: Files located at workflows/design-qa.md, <src>, and <file.html> are read or processed.
  • Boundary markers: None explicitly defined in the instructions.
  • Capability inventory: The agent utilizes python3 and node to run local scripts against these files.
  • Sanitization: Not explicitly mentioned, though the scope is restricted to design linting and headless rendering.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 05:32 PM
Security Audit — agent-trust-hub — design-qa