⭐️Introducing AG-UI; The Agent-User Interaction protocol 👾 Bring your AI agents into Frontend applications, and let them interact with users. Launching with day-0 integrations with LangGraph, CrewAI, Mastra, and AG2 - with more partnerships on the way 👀 github.com/ag-ui-protocol/ag… How it works & demo 👇

Hate that we have to do this, but PSA 🚨 This person is NOT affiliated with CopilotKit. We have been getting reports of sophisticated scams associated with this account. Including fake hiring. We have blocked him and suggest you do the same.

A2UI lets the agent build your UI at runtime from a catalog of React components you define. It composes the layout. Your components and CSS stay fixed. Full-stack demo: upload a PDF, ask a question, and the agent assembles the canvas from your catalog. Built with LangChain and CopilotKit. github.com/CopilotKit/Copilo…

AG-UI makes building agentic applications dramatically easier. Here's how it works. This is a model for a simple chatbot: User → LLM → Response But interactive agents that render UI, pause for approvals, and ask users for input need a much more complex model. When building these agents, a response from the LLM will include a series of state changes as the agent runs: • Agent started a task • Agent called a tool • Agent updated its state • Agent streams these tokens • Agent is waiting on a human • Agent is resuming the task The Agent-User Interaction Protocol (AG-UI) treats the LLM response as a stream of events rather than a text endpoint. In practice, here is what you get as an agent runs: 1. Lifecycle events so your UI knows where the agent is. 2. Text messages that stream tokens. 3. Tool calls so your UI can prefill a form with any required arguments. 4. State updates that keep your UI in sync with the agent. 5. Special events for human approvals, rich media, and custom needs. All of these events travel over standard transports (SSE, WebSockets, or plain HTTP) as JSON. As a result, you can build a frontend that stays in sync with the agent's progress without having to invent a custom process to make this happen. For example, building a human-in-the-loop workflow becomes an off-the-shelf component you can integrate rather than build from scratch. @CopilotKit is the creator of AG-UI, and you can use it when building frontend applications pretty much anywhere: • React • Angular • Vue • React Native • Slack • Teams • Discord • WhatsApp • Telegram Here is the link for you to check it out: github.com/CopilotKit/Copilo… Thanks to the CopilotKit team for partnering with me on this post.

👩‍🍳 CopilotKit is cooking at @aiDotEngineer World's Fair Stop by our booth, catch our joint demo with Oracle and come get the best merch in the conference

AWS integrates with AG-UI We've worked with @AWS to bring the AG-UI protocol natively into FAST, Bedrock AgentCore Runtime, and Strands Agents. @CopilotKit then extends AG-UI with - generative UI - shared state - human-in-the-loop Build generative UI↓ aws.amazon.com/blogs/machine…

Packed house last night at @SandboxVR for AI Engineer World's Fair week 🕶️ 200+ AI builders, VR, good bevs and even better chats. Thanks @nebiusai @tavilyai @CopilotKit for co-hosting!

day 2 at AIE ❕come say hi to @CopilotKit in the startup booths by the entrance

+1. SaaS tools especially CRMs should ship this into their Ask AI that they have already. Its easy to build with @CopilotKit. I put a quick demo on top of @attio data. @attio Ask AI is already great for follow ups and meeting prep something like this would be a great addition. You could probably do two modes: 1) Controlled: These are designed views for the questions you know you'd get asked. 2) Declarative: For everything else, the agent emits a UI schema and you render it from primitives on the fly.

AG-UI Repo: github.com/ag-ui-protocol/ag…

Connect any agent to any frontend with AG-UI An event-based, bidirectional protocol where the agent streams what it's doing and your UI renders it live. Start with: 𝚗𝚙𝚡 𝚌𝚛𝚎𝚊𝚝𝚎-𝚊𝚐-𝚞𝚒-𝚊𝚙𝚙

Someone shipped an open source alternative to Claude in Slack. 1 day after Anthropic launched it. It's called OpenTag. What you get: → Generative UI - renders charts, tables, and components inline inside Slack → Streaming replies with full thread context → Human-in-the-loop approval gates - agent waits for you before acting → One codebase. Slack, MS Teams, Discord, Telegram, WhatsApp. → MIT license - fully hackable The entire agent behavior is controlled by one system prompt. Rewrite it, you have a completely different agent. Built on CopilotKit's open SDK. 35k+ GitHub stars. Created the AG-UI protocol. Clone it, point it at your model and tools, and you own the whole stack.

Wheels down San Francisco for AI World's Fair! One week into my new chapter at @CopilotKit and already on the floor with the team. We're at the booth talking agentic frontends, the AG-UI protocol, and what it takes to ship agents to production. Come find us. Would love to meet you.

Introducing OpenBox for CopilotKit Give any AG-UI compatible agent runtime governance, human-in-the-loop approvals, and signed audit trails. Create production-ready agents ↓ docs.copilotkit.ai/cookbook/…

Bring any MCP App into your product, with: • Zero frontend code • Full interactivity (HTML/CSS/JS) • Secure sandboxing • Thread persistence See it in action ↓

MCP App Builder by @CopilotKit Pretty cool how much you can do with CopilotKit. You can build: - Lovable for [anything] - Cursor / Claude-like experiences - Any agent to Any frontend - All the different kinds of generative UI Everything running on the AG-UI protocol. Oh and it's all open source. Incredible github.com/CopilotKit/Copilo…

Render MCP Apps directly in your chat. Connect any server and its tools show up as interactive, sandboxed components. 𝚗𝚙𝚖 𝚒𝚗𝚜𝚝𝚊𝚕𝚕 @𝚊𝚐-𝚞𝚒/𝚖𝚌𝚙-𝚊𝚙𝚙𝚜-𝚖𝚒𝚍𝚍𝚕𝚎𝚠𝚊𝚛𝚎 Get started ↓ docs.copilotkit.ai/generativ…