build-awwwards-quality-sites
Pass
Audited by Gen Agent Trust Hub on Jul 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides detailed guidelines for web development using well-known, industry-standard libraries and services including GSAP, Three.js, and Iconify. The instructions emphasize original content creation, performance safeguards, and ethical asset provenance.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided reference evidence to extract high-level design traits. While this constitutes an ingestion surface for external data, the instructions explicitly mandate against the direct reproduction of external assets or code, serving as a functional safeguard against input-led behavior override.
- Ingestion points: User-provided reference evidence (SKILL.md, Section 1).
- Boundary markers: Not explicitly defined in the instructions.
- Capability inventory: Web code and asset generation (SKILL.md, Sections 1-6).
- Sanitization: Instructions to generate materially new identity and avoid tracing provide high-level constraints on input influence.
Audit Metadata