openui-forge-langchain

Pass

Audited by Gen Agent Trust Hub on May 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of several Node.js packages, including vendor-scoped packages such as @openuidev/react-ui, @openuidev/react-headless, and @openuidev/react-lang.
  • [COMMAND_EXECUTION]: Provides instructions to execute a CLI tool via npx @openuidev/cli for generating system prompts, which is a standard part of the library's integration workflow.
  • [PROMPT_INJECTION]: The skill implements a chat backend that processes user messages and passes them to an LLM, creating an indirect prompt injection surface.
  • Ingestion points: User messages are ingested via the POST request body in app/api/chat/route.ts.
  • Boundary markers: None identified; messages are mapped directly to LangChain message classes (HumanMessage and AIMessage).
  • Capability inventory: The skill facilitates streaming LLM output to a frontend UI and allows defining custom React components via the @openuidev/react-lang library.
  • Sanitization: No explicit sanitization, validation, or filtering of the user-provided message content is implemented in the provided backend code snippets.
Audit Metadata
Risk Level
SAFE
Analyzed
May 22, 2026, 04:07 PM