reviewer-plan
Pass
Audited by Gen Agent Trust Hub on Jul 3, 2026
Risk Level: SAFENO_CODECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious code or patterns were detected. The skill functions as a reviewer for architectural plans as described.
- [NO_CODE]: The skill contains instructional text only and does not execute scripts.
- [COMMAND_EXECUTION]: The skill retrieves project information using the
bdCLI tool. - [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection via issue descriptions. 1. Ingestion points:
bd showandbd listoutput inSKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Read-only access to codebase and issue tracker. 4. Sanitization: None detected.
Audit Metadata