phone-ui-demos
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user content (such as product feature lists, knowledge points, or scripts) to generate HTML-based animations.
- Ingestion points: User input is processed by the agent and placed into mobile UI components (chat bubbles, notifications) within the generated files (e.g., shot-1.html).
- Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions embedded within the user-provided data.
- Capability inventory: The generated output is executable HTML/JS code. The skill documentation in SKILL.md mentions a quality check process involving running a local Node.js script (node ../video-shot-demos/scripts/shot.js).
- Sanitization: There are no instructions for the agent to sanitize or escape user-provided text before interpolating it into the generated code, creating a potential surface for cross-site scripting (XSS) in the generated artifacts.
Audit Metadata