anti-slop-design
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and refactor user-provided UI code, design contracts, and rendered interface audits, which constitutes an attack surface for instructions embedded in processed data.
- Ingestion points: The skill analyzes user-supplied theme documentation, registry configuration files, design contracts, and live or rendered UI code (as specified in SKILL.md and references/implementation-recipes.md).
- Boundary markers: Instructions include clear scopes for implementation and verification, requiring the agent to "stop when the requested scope is covered" and "preserve existing designs," which serves as a natural constraint.
- Capability inventory: The skill is capable of generating and modifying frontend code (CSS, React, SVG). However, it explicitly forbids the addition of external dependencies or tests solely for the purpose of the skill, limiting the potential impact of an injection.
- Sanitization: The skill contains explicit instructions for the agent to maintain security boundaries, such as "never fabricate customer proof" and ensuring that "sensitive content and metadata must not be delivered merely to hide them on the client."
Audit Metadata