chat-ui
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads UI component registry files from 'ui.inference.sh', which is the vendor's official domain.
- [COMMAND_EXECUTION]: Instructions include the use of 'npx shadcn' to add remote components and 'npx skills' to install related CLI tools and UI packages.
- [INDIRECT_PROMPT_INJECTION]: The skill provides UI components like 'ChatMessage' which are designed to ingest and display untrusted data from users and AI assistants.
- Ingestion points: The 'content' property of the 'ChatMessage' component in 'SKILL.md'.
- Boundary markers: None explicitly defined in the UI component code snippets.
- Capability inventory: React component rendering only; no direct file system or system-level capabilities are exposed by these UI blocks.
- Sanitization: Standard React DOM rendering provides protection against XSS, but the skill does not explicitly detail content sanitization for AI-specific injection attacks.
Audit Metadata