writing-precis-reviewer
Pass
Audited by Gen Agent Trust Hub on Apr 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill enforces a structured document review process within a controlled environment.
- [SAFE]: Tool permissions are restricted to necessary functions (Read, Grep, Glob, Agent) for the primary agent.
- [SAFE]: The reviewer subagent is explicitly limited to read-only tools (Read, Grep, Glob) and is barred from using destructive tools like Write, Edit, or Bash, which mitigates risks during the processing of potentially untrusted document content.
- [SAFE]: File system interactions are limited to reading internal constraint references and writing status artifacts within the project's planning directory (".planning/").
- [SAFE]: No external network dependencies, remote code execution patterns, or credential management issues were identified.
Audit Metadata