frontend-design
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes local project source code, routes, and content as part of its primary workflow, which represents a potential surface for indirect instructions if the analyzed code is malicious.
- Ingestion points: Project source code, routes, components, tokens, and typography (SKILL.md).
- Boundary markers: Not present; the skill treats repository instructions and existing product behavior as authoritative without explicit delimiters.
- Capability inventory: File modification (build/redesign) and execution of local validation commands including linting, typechecking, and build processes (SKILL.md).
- Sanitization: No explicit sanitization or filtering of project-provided code or content is described.
- [EXTERNAL_DOWNLOADS]: The skill references technical documentation from MDN (developer.mozilla.org) regarding SVG and CSS properties. This is a well-known and trusted service for developer documentation.
Audit Metadata