architecture-review-gate
Pass
Audited by Gen Agent Trust Hub on Aug 13, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill incorporates a local Python script,
scripts/score_architecture.py, and instructs the agent to execute it for automated analysis of architecture documents. The script performs static regular expression scanning and uses only standard library modules.\n- [INDIRECT_PROMPT_INJECTION]: The skill has a data ingestion surface as it analyzes architecture specifications provided as external Markdown files. The instructions mitigate this risk through clear boundary markers, instructing the agent to treat all diagrams, reports, and claims as untrusted until they are corroborated with evidence.\n- [SAFE]: The skill demonstrates a security-conscious design. It contains no obfuscated code, remote downloads, persistence mechanisms, or unauthorized data access patterns. The functionality is consistent with its stated purpose of improving architecture quality and safety.
Audit Metadata