ai-core/chat-experience
Pass
Audited by Gen Agent Trust Hub on Sep 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill exclusively uses resources and libraries from the '@tanstack' organization, which is the verified author of the content. References to future model versions and example domains are used appropriately within a documentation context.
- [INDIRECT_PROMPT_INJECTION]: The skill demonstrates how to build an interface that processes user input and renders AI-generated content, which is a standard surface for indirect prompt injection. • Ingestion points: User messages sent via the
sendMessagehook and parsed by server-side routes. • Boundary markers: The examples utilize standard message arrays without additional explicit delimiter logic. • Capability inventory: The implementation supports MCP tool execution, file uploads, and multimodal processing. • Sanitization: The provided UI components render text and thinking parts directly; standard output validation is recommended for production environments.
Audit Metadata