a2ui
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill instructions include a 'Companion check' that directs the agent to verify installed recommended skills by checking specific directories (e.g.,
~/.claude/skills/) using thelscommand. This is a legitimate environment discovery task for setup convenience and dependency management. - [SAFE]: All external resources, including GitHub repositories and documentation sites, target official and well-known sources belonging to Google (
github.com/google/A2UI) or the primary project domain (a2ui.org). - [SAFE]: The skill provides explicit security guidance for implementers, recommending agent-side validation against JSON schemas and input sanitization for custom UI components.
Audit Metadata