Skills
skills/1nfsh/skills/chat-ui/Gen Agent Trust Hub

chat-ui

Warn

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADS
Full Analysis
  • EXTERNAL_DOWNLOADS (MEDIUM): The skill recommends installing components via npx shadcn@latest add https://ui.inference.sh/r/chat.json. This command downloads and integrates code from an external, third-party domain that is not on the trusted sources list.
  • EXTERNAL_DOWNLOADS (MEDIUM): The 'Related Skills' section suggests using npx skills add to install additional packages from the inference-sh namespace, which involves fetching external code.
  • INDIRECT_PROMPT_INJECTION (LOW): The ChatMessage component is designed to render untrusted data from AI responses or users.
  • Ingestion points: content prop in ChatMessage (SKILL.md).
  • Boundary markers: None present in the example code.
  • Capability inventory: React UI components; potential for XSS if combined with unsafe markdown rendering (suggested in related skills).
  • Sanitization: Not explicitly shown; depends on implementation details of the imported registry blocks.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 17, 2026, 06:21 PM
Security Audit — agent-trust-hub — chat-ui