CopilotKit🪁

⭐️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… How it works & demo 👇

We'll cover: - How our agent self-improvement works - Turning user interactions into learning signals - Prompt Mutation and automated improvement - Per-user adaptation and personalized behavior - Why this replaces traditional fine-tuning - How to get early access Hosted by @nathan_tarbert with @MikeRyanDev!

Your Agent should have self-learning Join our Webinar for a deep dive into how you can let agents learn from real user interactions through Continuous Learning from Human Feedback. This Thursday, 12pm PST RSVP: luma.com/copilo-f5ii

Gartner recognizes CopilotKit as a Cool Vendor Specifically for AG-UI, the open protocol we built to connect AI agents to user-facing apps. Thank you to the team and everyone building on @CopilotKit 🪁 The future of UI is AI

WorkPods agents can now run on React Native. Aside from generating text, the agent can now drive real native components on mobile, similar to dynamic dashboards and canvas on the web. All powered by @LangChain . @CopilotKit @LangChain @swmansion @Baconbrix

Spent the weekend tinkering on @danshipper Codex native app idea. Here is a basic demo using @CopilotKit on top of @attio CRM data. Attio here is the data layer/source of truth. CopilotKit generates the UI on the fly Codex becomes the agentic workflow layer through its native browser. The important bit is that the generated UI can hand context back to Codex. Feel like this version where AI generates UI on the fly, will be better than computer use. With browser use the agent still has to operate through whatever UI the SaaS app gives you. Here the UI can be generated for the exact task and context in front of you. Since it all runs inside Codex you also get your own skills, plugins, tools & context handoff built in. It goes back to the idea that Codex & Cowork become the layer for knowledge work. The app is no longer the surface. The UI is just whatever the task needs.

Your agent describes the UI. CopilotKit renders it inline. Ask for the weather, a live weather card shows up in chat. That's A2UI. Open-JSON-UI, MCP Apps and custom specs render the same way. Learn more about it 👇 docs.copilotkit.ai/built-in-agent…

AWS just made building production-ready AG-UI Web Apps a one-command experience. You can scaffold a full AI app with: - Strands agent on AgentCore - AG-UI protocol as the streaming layer - @CopilotKit on the frontend - infra and auth ready Try it out → awslabs.github.io/nx-plugin-for-…

AWS 🤝 AG-UI The Nx Plugin for @AWS now scaffolds AG-UI Web Apps in TypeScript or Python. One command gives you: Strands agent on AgentCore, AG-UI on the wire, CopilotKit on the frontend, auth + infra. Try it → awslabs.github.io/nx-plugin-for-…

Want to build your own? CopilotKit is open source (MIT) — React, the runtime, AG-UI, and now React Native, all in one repo: github.com/CopilotKit/Cop… Start with `npm i @copilotkit/react-native`

CopilotKit now runs in React Native. The same hooks that drive agent-native UI on the web: useFrontendTool, human-in-the-loop, generative UI rendered in chat — now ship for iOS and Android via @copilotkit/react-native. Agents that drive real components, on mobile. Demo + repo 👇

The shared state loop 🧵, start to finish: Your component subscribes to the agent's state. Run kicks off and the agent receives a snapshot of exactly what the UI shows. Mid-run, the agent updates a value. That delta streams to the frontend immediately, not at the end of the run, and React re-renders with it. User changes something? It's in the agent's state by its next turn. No polling, no parsing progress out of chat messages. The state object is the channel. That's shared state in CopilotKit Docs: docs.copilotkit.ai/shared-state

Registration is still open @unicorn_mafia 🦄 luma.com/a2a-and-genera…

London’s FIRST Protocol and Gen UI Hackathon Happening tomorrow! With over 300+ registered 🎉 Come build agents that either 1. communicate with other agents 2. generate their own UI Londons best will be there with @sofiiiiiasz. Sign up below👇 Sponsored by CopilotKit, @Google, @Redisinc, @SierraPlatform, @Linkup_platform

We recently had a Webinar on this, check out the recording 👉 youtube.com/watch?v=qA2L4E…

Sign up for early access: go.copilotkit.ai/beyond-the-web…

Slack 🤝 AG-UI Protocol @SlackHQ is now a first-class AG-UI frontend with: → Streaming replies → Tool calls → Human in the Loop approvals → Full thread context Same agent you built for the web, now living where your users actually work.

Daytona 🤝 CopilotKit 🎉 Excited to release our new cookbook In our first guide, your @CopilotKit Built-in Agent can run code inside a Daytona sandbox and return the result to your chat. Shoutout to @ivanburazin, @ivanda_mislav and the Daytona team for making this happen alongside Mark Fogle, @tyler_cpk and @sofiiiiiasz. Looking forward to more collaborations 🪁

"Build me a Snake game in Vite + React" → game runs inside the chat. "Make it red themed" → the running game updates live. We worked with @CopilotKit to build a generative-UI coding agent backed by @daytonaio sandboxes. Every tool call streams as a live card. So does the running app.

Great collab. The @CopilotKit team builds solid stuff and wiring it to Daytona sandboxes felt like an unlock.

@eli_discovers @daytonaio 🧑‍🍳

@CopilotKit @daytonaio The team really cooking here

Daytona 🤝 CopilotKit Your agent can now run code & preview outputs Drop in a @daytonaio sandbox tool and our Built-in Agent executes Python, TS, or JS in an isolated cloud sandbox, then returns the result back to your chat. Live demo: …ona-runcode-production.up.railway.app