design-review

Pass

Audited by Gen Agent Trust Hub on Apr 2, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates entirely within the local filesystem, reading design briefs and project code to generate a critique. No external network connections, sensitive file access (such as credentials), or privilege escalation attempts were detected.
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes untrusted content from the codebase and design briefs.
  • Ingestion points: Reads .design/*/DESIGN_BRIEF.md and various project source files.
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to treat the content of these files as data rather than instructions.
  • Capability inventory: The skill uses file-read capabilities and writes the output to DESIGN_REVIEW.md.
  • Sanitization: No sanitization of the input file content is implemented before processing. This risk is inherent to the intended purpose of code review and is mitigated by the localized nature of the tool.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 2, 2026, 08:45 AM