krt-interface-warden
Pass
Audited by Gen Agent Trust Hub on Oct 30, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides instructional guidance for UI/UX design and implementation. It does not include any executable scripts, command-line operations, or persistence mechanisms.
- [EXTERNAL_DOWNLOADS]: The skill contains informational links to established design systems and academic research (Material Design, U.S. Web Design System, GitHub Primer, Atlassian, and arXiv). These are well-known documentation sources and do not involve the execution of remote code or the download of untrusted assets.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external data, such as UX contracts from a 'Guardian' skill and critique briefs from an 'Inquisitor' skill, which it then uses to generate frontend code. While this creates a surface for indirect prompt injection, the skill mitigates this by requiring a 'Composition Brief' to rationalize all implementation decisions before generating code, providing a clear functional boundary between input and output.
- [DYNAMIC_EXECUTION]: The skill generates JSX and CSS code at runtime based on defined layout archetypes and surface metaphors. This is the intended primary purpose of the skill and is handled through standard large language model code generation without unsafe deserialization or runtime compilation of untrusted scripts.
Audit Metadata