AI Briefing
KOSign in

AI SDK fx adapter moves harness execution to host process

·2026.10.27 09:00

Key point

The rewrite uses libfx to eliminate sandbox bridges and port exposure while keeping model credentials on the host.

Details

  • The fx adapter for the AI SDK’s HarnessAgent now runs the harness and model requests in the same process as the application.
  • Built-in filesystem and shell operations remain isolated in the sandbox, while model credentials stay on the host rather than being forwarded into the sandbox.
  • The adapter uses libfx, the TypeScript library for fx, removing the need to install fx inside the sandbox, run a sandbox bridge, or expose a sandbox port.
  • New features include built-in tool filtering, mid-turn steering, structured output via schema matching, and image tool results returned through toModelOutput().
  • Sessions can be saved and resumed in a new host process; users must update @ai-sdk/harness-fx to the latest version to use the new adapter.
typescriptai-sdksandboxharness-agentfx-adapterlibfx
Read originalShare on Threads

This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.

Related

Vercel BlogVercel Sandbox adds stdin support for running commandsPyTorchKRVercel durable-harness supports log-based turn resumption on function failureVercel BlogVercel Plugin adds API guidance and specialist agents to OpenCodePyTorchKRwhatsapp-mcp: WhatsApp conversation search MCP server released without sending functionalityElevenLabsHow to integrate ElevenLabs TTS APIs: A step-by-step guide for Python and TypeScriptGeekNewsCopilotKit Releases OpenIntelligentUI, an Open-Source Alternative to ChatGPT Intelligent UI