prototype-feature
Pass
Audited by Gen Agent Trust Hub on Aug 3, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is designed for local UI prototyping. It contains explicit instructions to avoid backend calls, schema changes, or external network activity, effectively isolating the generated code.
- [PROMPT_INJECTION]: The skill processes natural language feature requests (SKILL.md). While this ingestion of untrusted data is a surface for indirect prompt injection, the risk is mitigated by instructions mandating adherence to existing application patterns and local-only data. Ingestion points: feature requests in description and Build section. Boundary markers: absent. Capability inventory: filesystem access for reading and writing UI components. Sanitization: absent.
Audit Metadata