NYC
skills/assistant-ui/skills/runtime/Gen Agent Trust Hub

runtime

Pass

Audited by Gen Agent Trust Hub on Feb 15, 2026

Risk Level: LOW
Full Analysis
  • [EXTERNAL_DOWNLOADS] (LOW): References legitimate NPM packages including @assistant-ui/react, openai, zustand, and react-redux. These are well-known libraries and consistent with the skill's stated purpose of UI state management.
  • [COMMAND_EXECUTION] (SAFE): No instances of shell command execution or unsafe subprocess calls. Code snippets are limited to React and frontend logic for managing AI chat interfaces.
  • [DATA_EXFILTRATION] (SAFE): Network operations shown in examples (such as fetch to /api/chat) are standard for chatbot functionality. There are no attempts to access sensitive local file paths like SSH keys or environment secrets.
  • [INDIRECT_PROMPT_INJECTION] (LOW): The skill documents how to handle message and tool-call data. While the resulting application will process untrusted external content, these instructions provide standard implementation patterns and do not introduce malicious bypasses or unsafe interpolation logic.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 15, 2026, 08:41 PM