overdrive
Pass
Audited by Gen Agent Trust Hub on Aug 15, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends using several well-known, industry-standard frontend libraries to achieve advanced visual effects and performance, including Three.js, GSAP, Framer Motion, and TanStack Virtual.
- [DATA_EXPOSURE]: The instructions mention the use of device-level APIs such as geolocation and ambient light sensors for interactive UI. The skill explicitly mandates that these features must only be used with explicit user permission and should be applied sparingly.
- [INDIRECT_PROMPT_INJECTION]: The skill handles untrusted data when rendering large datasets in tables or charts. It mitigates risks by requiring browser automation for visual verification and iterative design proposals, ensuring human oversight of the rendered output.
- Ingestion points: User-supplied datasets intended for rendering in tables, charts, or visualizations (SKILL.md).
- Boundary markers: Not explicitly defined for the data ingestion phase, but the skill follows a strict Context Gathering Protocol.
- Capability inventory: Utilizes standard browser rendering APIs (WebGL, WebGPU, Canvas, DOM) and client-side processing (Web Workers, WASM).
- Sanitization: The skill focuses on presentation; standard browser-based DOM/Canvas security models apply to the rendering of external content.
Audit Metadata