site-architecture
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill serves as a guide for information architecture and does not perform any automated actions, command executions, or network requests.
- [SAFE]: No obfuscation or hidden instructions were found across the documentation and reference files.
- [SAFE]: The practice of reading local context files like
product-marketing.mdis a standard way for agents to understand the user's project scope and does not constitute unauthorized data access in this context. - [INDIRECT_PROMPT_INJECTION]: The skill identifies a potential attack surface by ingesting external data from files like
.agents/product-marketing.md. However, because the skill has no code execution capabilities or external tool access (beyond standard text/diagram generation), this risk factor does not elevate the verdict.
Audit Metadata