Threadplane
PricingTalk to Us
DocsRunCode is unavailable because this page has no workspace capability.API is unavailable because this page has no workspace capability.
Choosing an adapter

Documentation

Start building with Threadplane

Streaming agent interfaces with runtime adapters, a shared Agent contract, and a drop-in chat surface. Every package is MIT-licensed, with optional production assurance and delivery services for teams that want expert support.

1Pick your backend

LangGraph

LangChain

For LangChain & LangGraph backends.

npm i @threadplane/chat @threadplane/langgraph @langchain/core @langchain/langgraph-sdk marked
Adapter quickstart →

AG-UI

AG-UI

For CrewAI, Mastra, Pydantic AI, Strands, and more.

npm i @threadplane/chat @threadplane/ag-ui @ag-ui/client @ag-ui/core marked
Adapter quickstart →

Not sure which to use? Choosing an adapter → Want the drop-in UI first? Chat quickstart → Running a non-LangGraph backend? Agent runtimes → Building on the Deep Agents framework? Deep Agents →

2Generative UI

A2UI

Google

Agent-to-UI protocol — the agent streams and updates surfaces over the conversation.

Get started →

json-render

Vercel

Render a fixed JSON spec into your own Angular components. You own the schema.

Get started →

Which fits my use case? json-render vs A2UI →

3Chat UI

Chat

Threadplane

Drop-in chat components — message list, input, streaming, tool calls, interrupts, subagents. Renders A2UI & json-render surfaces inline.

Supporting libraries

Middleware

JS & Python client-tool routing

Looking for something specific?