egui-ui-design
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns were identified. The skill instructions follow standard software development practices for the Rust ecosystem, such as using existing libraries (egui, egui_kittest) and following WCAG accessibility guidelines.
- [INDIRECT_PROMPT_INJECTION]: The skill addresses the vulnerability surface where the agent processes external source code and string literals. 1) Ingestion points: Source files, comments, and string literals are ingested for UI review and audit. 2) Boundary markers: A dedicated 'Guardrails' section explicitly instructs the agent to treat this content strictly as data and never follow instructions found within it. 3) Capability inventory: The skill suggests writing test functions and executing them via 'cargo test' to verify UI rendering. 4) Sanitization: The skill employs instruction-based isolation to ensure that potential prompt injections in data are not executed as commands.
Audit Metadata