micro-interaction
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is focused on UI animation development using industry-standard libraries like Framer Motion and modern CSS features. The instructions and code recipes provided are consistent with its stated purpose and present no malicious behavior.
- [COMMAND_EXECUTION]: The skill references a verification workflow utilizing standard development tools like playwright (via npx) and local helper scripts (scripts/seek-shot.sh, scripts/contact-sheet.sh) for capturing UI states. These tools are intended for local development and testing of the generated animations.
- [INDIRECT_PROMPT_INJECTION]: A vulnerability surface exists where instructions embedded in user-provided UI design requests could influence the generated HTML/JS code. 1. Ingestion points: User descriptions of UI interactions (e.g., 'add a hover effect') defined in the frontmatter description. 2. Boundary markers: The skill provides specific animation principles (durations, easing) and Framer Motion recipes that guide the agent's output. 3. Capability inventory: The skill generates executable HTML/JS code and uses playwright and local scripts for verification. 4. Sanitization: No specific sanitization of user input is mentioned.
- [EXTERNAL_DOWNLOADS]: The skill references the esm.sh CDN for importing React and Framer Motion in standalone demos. esm.sh is a well-known service for serving JavaScript modules. The skill also includes a promotional link to the author's website (iart.ai) in the documentation footer.
Audit Metadata