animation-craft
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references numerous external resources for research and documentation purposes. These include academic papers from Tactual Labs and official developer documentation from Apple, Google (web.dev), Mozilla (MDN), and Microsoft Applied Sciences. These are well-known and trusted sources for technical guidance.
- [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest external data in the form of "MOTION SPEC" or "BEAT SHEET" blocks to generate UI animation code. This design presents a potential surface for indirect prompt injection where instructions could be embedded in the data to influence the agent's behavior.
- Ingestion points: User-provided Markdown blocks or external files containing motion specifications described in SKILL.md.
- Boundary markers: The skill relies on natural language headers (e.g., "MOTION SPEC", "BEAT SHEET") to identify data sections rather than strict structural or cryptographic boundaries.
- Capability inventory: The skill produces executable code in CSS, JavaScript, SwiftUI, and React Native (Reanimated/Skia), which may be written to the filesystem or executed by the agent's environment.
- Sanitization: The skill does not explicitly define sanitization or filtering protocols for the input data before processing it into generated code.
Audit Metadata