book-project-architect
Pass
Audited by Gen Agent Trust Hub on Jul 18, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is purely instructional, focused on diagnosing book concepts and routing them to other specialized skills. It follows a structured workflow without accessing sensitive system resources.
- [NO_CODE]: There are no scripts, binaries, or shell commands associated with this skill. It relies entirely on Markdown-based instructions and reference files.
- [PROMPT_INJECTION]: No evidence of jailbreak attempts or instruction overrides. The skill includes a dedicated 'risk-routing' reference file that specifically instructs the agent to escalate sensitive topics (health, law, finance) to a specialized review process, which enhances the safety profile of the agent.
- [DATA_EXFILTRATION]: The skill does not perform network operations or access sensitive local files. It only references its own internal documentation files.
Audit Metadata