site-architecture
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed strictly of instructional Markdown content and reference templates. It contains no executable scripts, binaries, or active code components.
- [INDIRECT_PROMPT_INJECTION]: The skill includes instructions to read local configuration files for context, which is a potential data ingestion surface. 1. Ingestion points: The skill (SKILL.md) reads .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md to gather business context. 2. Boundary markers: There are no specific delimiters or ignore-instructions markers defined for the ingested content. 3. Capability inventory: The skill only performs informational tasks (planning hierarchies, designing navigation) and does not have access to subprocesses, network operations, file-writing, or eval/exec tools. 4. Sanitization: No sanitization or validation of the ingested context files is mentioned.
Audit Metadata