ai-architect-review
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill involves analyzing external codebase data, which presents a surface for indirect prompt injection. The instructions proactively address this by teaching the agent to identify missing trust boundaries and ensure that untrusted content (retrieved documents, API responses) is not treated as instructions.- [COMMAND_EXECUTION]: The skill uses the
rg(ripgrep) utility to search for provider SDK imports during code analysis. This is a legitimate diagnostic use of local command-line tools.
Audit Metadata