emilkowalski-motion
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill operates by ingesting and modifying existing UI artifacts such as HTML elements and React components. This ingestion creates a theoretical attack surface where malicious instructions embedded in the processed artifacts could attempt to influence the agent's behavior.
- Ingestion points: Referenced in the workflow as current HTML, component, or selected page elements.
- Boundary markers: Includes specific instructions to keep copy, data, and layout intent intact, providing a logical constraint on modifications.
- Capability inventory: Generates CSS keyframes, JavaScript observers, and GSAP animation sequences.
- Sanitization: None explicitly defined for the input artifact content.
Audit Metadata