feature-shape
Pass
Audited by Gen Agent Trust Hub on Sep 13, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists exclusively of high-level guidance for the AI agent to facilitate architectural discussions and documentation. No malicious patterns were identified.
- [INDIRECT_PROMPT_INJECTION]: The skill instructions involve analyzing repository content and external documentation. While this creates a surface for indirect prompt injection from untrusted files or websites, the skill lacks dangerous tools (such as arbitrary code execution or network exfiltration) that would make such an injection actionable.
- Ingestion points: Repository analysis and external documentation checks mentioned in
SKILL.md. - Boundary markers: Not explicitly defined in the prompt.
- Capability inventory: None detected; the skill primarily produces text-based reasoning and documentation.
- Sanitization: Not applicable as no external data is executed or passed to shell commands.
Audit Metadata