prd-to-architecture

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill serves as an instructional template for system design. It operates entirely on local project files within a defined directory structure (docs/ai-native/).
  • [SAFE]: No network access, remote code execution patterns, or external downloads were identified in the instructions or metadata.
  • [SAFE]: The skill does not handle sensitive information or credentials; it focuses on generating architectural documentation and data models based on user-provided PRDs.
  • [SAFE]: The project follows a strict 'Source Loading Protocol' which ensures the agent reads from local files rather than relying on potentially truncated chat history.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 04:18 AM
Security Audit — agent-trust-hub — prd-to-architecture