Anmol Baranwal and Nathan TarbertAugust 4, 2026
Introducing Channels SDK - Bring Any Agent to Any Channel
We're introducing the Channels SDK, an open source library that brings any AG-UI agent into Slack, Microsoft Teams, and every other chat platform, from one codebase. Write your agent once and it runs natively on each channel, with generative UI, tools, MCP servers, human-in-the-loop approvals, file uploads, and memory that follows the user across channels.
Subscribe to the newsletterGet notified of the latest news and updates.
Sam Julien September 26, 2026
AG-UI for Product Managers: Building Agents People Can Work WithA practical guide to AG-UI for product managers: shared context, user control, agent interactions, and learning from real product outcomes.
Anmol Baranwal and Eli BermanSeptember 25, 2026
AG-UI Protocol now has a first-class .NET SDKMicrosoft just contributed a first-class .NET SDK to AG-UI, so any .NET backend can speak the protocol directly. C# support used to live only inside Microsoft Agent Framework, which now depends on these packages. The .NET SDK has reached 1.0 on NuGet alongside AG-UI 1.0, and your endpoint works with any AG-UI client, from a React app to Slack or Teams.
Anmol Baranwal and Eli BermanSeptember 10, 2026
New and improved Inspector: DevTools for AI Agents, now with Automatic LearningWe've rebuilt the Inspector, DevTools for AI agents that runs inside your app. See every AG-UI event as it happens, jump straight to whatever broke, fork any saved conversation into an isolated Playground, and let your agent learn from real conversations with CopilotKit Intelligence connected. On by default in every development build, across React, Vue, and Angular.
Nathan Tarbert September 10, 2026
Connect a single Agent to 3 surfaces: React, Slack, TeamsOne agent, three front doors. See how AG-UI ships the same agent to Slack, Teams, and React — with generative UI, human in the loop, and shared state.
Eli Berman September 3, 2026
Introducing WebMCP for CopilotKitYour frontend tools shouldn't be locked to one copilot. WebMCP support is now live in CopilotKit: every tool you register with useFrontendTool is discoverable and callable by any WebMCP-compatible browser agent — ChatGPT Atlas, Comet, Dia, and whatever comes next. Opt-in per tool, auto-register on mount, works across React, Vue, Angular and more. Open source and requires just one line of config.
Anmol Baranwal and Sofía Sánchez-ZárateSeptember 2, 2026
Slack Agents Night: OpenTag running on the Channels SDKWe hosted Slack Agents Night with WorkOS in San Francisco, where Sofía from our team demoed Kite and OpenTag running on the Channels SDK. Kite drafted a real PR from a Slack thread, rendered native Block Kit components, and ran alongside Antigravity in the same conversation. Here's the full recording plus a breakdown of how the Channels SDK puts any AG-UI agent into Slack and Teams.
Eli Berman September 1, 2026
OpenBot passes 3,500 GitHub stars! Here's what's new in v0.0.5OpenBot just crossed 3,500 GitHub stars, and v0.0.5 has tons of new features. Bots can now answer from Google Drive and Notion, hand off to each other, run on a schedule, and route untagged messages on their own. You can take the wheel of a Bot's browser mid-task, ship skills for smarter tool use, sign in with enterprise SSO, deploy via Helm on AWS, dry-run boundary rules, and lean on a wide security hardening pass. Open source and self-hostable! Check it on GitHub.
Anmol Baranwal and Nathan TarbertAugust 24, 2026
Mastra is now fully AG-UI compatibleMastra is now fully AG-UI compatible, so the entire CopilotKit feature set runs on any Mastra agent, local or remote - including generative UI, shared state that streams live, interrupts, background tasks as live activity cards, frontend tools, subagents, multimodal, voice, reasoning and more. A2UI surfaces heal themselves, and generation runs in a dedicated sub-agent. Works across React, React Native, Next.js, and Angular on the web, plus Slack, Teams, and other messaging platforms via the Channels SDK. Everything is open source.
Nathan Tarbert and Anmol BaranwalAugust 19, 2026
React Native: Your Agent, inside Your Mobile AppYou can now bring any AG-UI agent into an iOS or Android app. The React Native SDK renders into native components instead of a webview, with the same agent, runtime, and tool protocol you already use on the web. Frontend tools run on the device next to your state and platform APIs, generative UI renders inline in the chat, and approval flows, interrupts, and threads carry over unchanged.
Jerel Velarde and Eli BermanAugust 18, 2026
CrewAI is now fully AG-UI compatibleHere are the newest updates and resources to start building state-of-the-art frontends for any CrewAI agent. ag-ui-crewai 0.3.0 brings CrewAI to full AG-UI support: streamed reasoning, multimodal input, human-in-the-loop interrupts, thread persistence and checkpointing, multi-agent attribution, generative UI over A2UI, and full feature parity for CrewAI's Conversational Flows.
Eli Berman August 14, 2026
aimock: Deterministic Mock Infra for AI Appsaimock is open-source mock infrastructure for every AI service your app talks to — LLMs, MCP tools, vector DBs, image models, TTS, and more. Write a JSON fixture, run npx aimock, point your API URL at localhost. Tests run offline in milliseconds and are completely deterministic, so zero tokens burned. Now at 1M+ weekly installs and 700+ stars, quietly powering CI at OpenClaw, TanStack, Mastra and the AG-UI protocol. From the team at CopilotKit.
Nathan Tarbert and Anmol BaranwalAugust 13, 2026
Building a Production-Ready Slack Agent with Channels SDKWatch Mike Ryan (CopilotKit Engineer) build a production-ready Slack agent, using Mastra, end to end with the Channels SDK. It's the open-source library that brings any AG-UI agent into Slack, Teams, Discord, WhatsApp, and more chat platforms. Learn how an agent goes from a message in a channel to a real response, with tools, rich native messages, and durable memory across a thread.
Eli Berman August 12, 2026
Claude Managed Agents are now AG-UI compatibleAnthropic just shipped a full AG-UI integration for Claude Managed Agents (CMA), so you can drop a Managed Agent into any AG-UI-compatible frontend without gluing together your own transport, state sync, and UI primitives. That means React, React Native, Next.js, and Angular on the web, plus Slack, Teams, and other messaging surfaces via the Channels SDK — all with generative UI, streaming, tool calls, and stateful sessions. Everything is open source. Here's where to start and how to ship.
Rainer Hahnekamp and Murat SariJuly 23, 2026
Introducing CopilotKit for AngularWe're introducing Angular support for CopilotKit. Bring any AG-UI agent into your Angular app with pre-built chat or a fully headless setup, Generative UI, shared state, human-in-the-loop and threads, all on Angular's own patterns. Soverius AI maintains it upstream, in collaboration with CopilotKit.
David McKay July 20, 2026
Auto-Tune: Stop Writing the Prompt, Train the ModelProduction AI teams often rely on massive system prompts to force general-purpose models into narrow, repetitive jobs—but instruction-following degrades as those prompts grow. This post argues for “auto-tune”: a simple, continuous system that learns from production traffic, trains smaller specialist models, and routes work to the right model automatically.
Anmol Baranwal and Nathan TarbertJune 30, 2026
Build Generative UI Agents on Amazon Bedrock AgentCore with AG-UI and CopilotKitAmazon Bedrock AgentCore now runs AG-UI agents natively, so auth, scaling, and session isolation come managed. We build one agent that renders charts inline, syncs a todo canvas both ways, and pauses for human input. Pick Strands or LangGraph at deploy time, and the same frontend runs against either.
Jordan Ritter June 12, 2026
Switching to Fable 5: The Tuesday That Cost $22,000CopilotKit swapped their coding agents to a new model (fable-5) on a Tuesday, and by Wednesday had run up a ~$22,000 Anthropic bill. No bug or runaway script caused it — five engineers doing normal work triggered it, because the new model interpreted their English-prose behavioral rules ("keep sub-agents small," "converge to zero") far more loosely than older models did. The result was four simultaneous drift modes: bloated sub-agents, runaway review loops, exploding fanout, and oversized single turns.
What kept it from being far worse was their "skills" system — small, named instruction packs that agents auto-load to encode team conventions. Their instrumentation caught the runaway in 12 hours instead of 12 days, isolated worktrees contained the blast radius, and the fix lived at the rules layer (swapping vague adjectives for hard numeric budgets and machine-checkable gates) rather than in application code.
The takeaway: if you let AI agents do real work, you need a guardrails layer that encodes desired behavior separately from the agents — small, composable, and faster to change than a model. Because models will change, and the team with a rules layer pays $22K and writes a blog post; the team without one pays more and writes a press release.
Nathan Tarbert June 12, 2026
Build AI Agents That Live Inside Your App — CopilotKit ExplainedCopilotKit is a complete set of building blocks for developers and teams who want to add agents to their app and assemble, debug, and ship the interface layer between their agents and users. It's also framework-agnostic and supports any LLM and every popular agent framework. There are two options for chat. Pre-built components, if you want something already built or Headless UI, which allows anyone to build custom agent interfaces. 
Anmol Baranwal and Nathan TarbertMay 20, 2026
CopilotKit - The Complete Frontend Stack for AI AgentsCopilotkit is the open source frontend stack for AI agents. Learn about architecture, chat components, hooks, generative UI, persistent memory, debugging tools, and 13+ agent framework integrations for building Agentic UIs.
Jordan Ritter May 18, 2026
TanStack Supply Chain Attack and How to Lock Down GitHub ActionsHow TanStack got hit through GitHub Actions, the 8 vulnerabilities we found auditing 20 repos the next day, and the playbook we used to fix them all in 4 days.
Anmol Baranwal and Nathan TarbertMay 14, 2026
CopilotKit Enterprise Intelligence Platform: The Persistence Layer for Agentic ApplicationsThe CopilotKit Enterprise Intelligence Platform brings persistent memory to agent apps across sessions and devices, with continuous learning from real usage.
Nathan Tarbert April 24, 2026
Build a Multi-Agent Telecom Support System with CopilotKit & LangGraphIn this tutorial, we’ll build a multi-agent support system using CopilotKit and LangGraph. In this system, different agents coordinate to resolve customer queries.
Anmol Baranwal and Tyler SlatonApril 23, 2026
Generative UI Spectrum: How Agents Now Ship Their Own InterfacesThe interface used to ship with the app. Now it ships with the agent. A walkthrough of the Generative UI spectrum (controlled, declarative, open-ended), the AG-UI protocol underneath, and the shared agent state that makes it all work.
Anmol Baranwal and Nathan TarbertApril 17, 2026
A2UI v0.9: What's New in Google's Generative UI SpecGoogle launched A2UI v0.9, their open protocol for Generative UI. It ships prompt-first generation, bidirectional messaging, modular schemas, and a new Agent SDK. Here's what changed, how to wire any agent in 5 steps, and how to use it with any agent framework via AG-UI Protocol.
Eli Berman April 10, 2026
Join CopilotKit at Google Cloud Next 2026We’ll be at Google Cloud Next 2026 in Las Vegas!
Google Cloud Next is the biggest cloud event of the year, and CopilotKit is showing up in a big way. CEO Atai Barkai has been invited to the stage in an official session with Google Cloud, we're co-presenting one of the best sideline events of the week, and our team will be on the ground meeting with developers and companies building the next generation of AI-native applications.
Anmol Baranwal and Nathan TarbertApril 9, 2026
AIMock: One Mock Server For Your Entire AI StackAIMock mocks your entire agentic stack from one config file: LLMs, MCP tools, A2A agents, vector databases, search, rerank, and moderation. Record real API responses as fixtures, detect provider changes daily before they reach production, and inject failures to prove your app handles them. Zero dependencies. Open Source.
Anmol Baranwal and Nathan TarbertMarch 24, 2026
AWS Announces Dedicated AG-UI Endpoint in AgentCore and FAST Template for Building Fullstack AgentsAWS and CopilotKit release a dedicated AG-UI endpoint in AgentCore and a new FAST template pattern with Generative UI, shared state, and human-in-the-loop flows out of the box.
Anmol Baranwal and Nathan TarbertMarch 12, 2026
Reusable Agents Meet Generative UIsOracle, Google, and CopilotKit have jointly released an integration that standardizes how AI agents are defined, how they communicate with frontends in real time, and how they describe the UI they require.
The integration connects three distinct layers. Oracle's Open Agent Specification (Agent Spec) provides a framework-agnostic way to define agent logic, workflows, and tool usage once and run it across compatible runtimes. AG-UI handles the live interaction stream between the agent and the frontend, keeping tool progress, state updates, and user interactions synchronized while the agent is executing.
A2UI, developed by Google, allows agents to describe the UI they need - forms, tables, multi-step flows - as structured JSONL, which CopilotKit then renders automatically inside the host application.
Previously, each of these layers required custom implementation per project. This release establishes a shared contract across all three, meaning agent developers can define the agent once, expose a standardized interaction stream, and have the frontend render structured UI surfaces without writing custom wiring for each tool or workflow.
The practical impact is reduced integration friction across the ecosystem - agent runtimes and frontend clients that implement these standards can interoperate without lock-in to a specific framework or vendor.
Anmol Baranwal and Nathan TarbertJanuary 29, 2026
The Developer's Guide to Generative UI in 2026AI agents have become much better at reasoning and planning. The UI layer has mostly stayed the same, and it is holding back the experience.
Most agent experiences still rely on chat, even when the task clearly needs forms, previews, controls, or step-by-step feedback.
Generative UI is the idea that allows agents to influence the interface at runtime, so the UI can change as context changes. This is usually done through UI specs like A2UI, Open-JSON-UI, or MCP Apps.
We'll break down Generative UI, the three practical patterns, and how CopilotKit supports them (using AG-UI protocol under the hood).
Anmol Baranwal and Nathan TarbertJanuary 22, 2026
Bring MCP Apps into your OWN app with CopilotKit & AG-UIToday, we are excited to announce CopilotKit’s support for MCP Apps. Now, MCP servers can finally ship an interactive UI that works out of the box in real agent applications.
Anmol Baranwal and Nathan TarbertJanuary 20, 2026
How to build a Frontend for LangChain Deep Agents with CopilotKit!LangChain recently introduced Deep Agents: a new way to build structured, multi-agent systems that can plan, delegate, and reason across multiple steps.
It comes with built-in planning, a filesystem for context, and subagent spawning. But connecting that agent to a real frontend is still surprisingly hard.
Today, we will build a Deep Agents powered job search assistant and connect it to a live Next.js UI with CopilotKit, so the frontend stays in sync with the agent in real time.
Nathan Tarbert and Eli BermanDecember 18, 2025
Oracle adopts AG-UI protocol for Agent SpecOne of the most complex problems in agentic systems today isn’t defining intelligent agents, it’s connecting those agents to real, reliable user experiences. Oracle’s Open Agent Specification (Agent Spec) and AG-UI now work together to solve that gap.
Bonnie and Nathan TarbertDecember 17, 2025
Build with Google's new A2UI Spec: Agent User Interfaces with A2UI + AG-UIIn this guide, you will learn how to build full-stack agent-to-user interface (A2UI) agents using the agent-to-agent (A2A) protocol, the AG-UI protocol, and CopilotKit. 
Nathan Tarbert December 15, 2025
AG-UI and A2UI Explained: How the Emerging Agentic Stack Fits TogetherThe agentic ecosystem is evolving fast — and with new standards appearing almost weekly, it’s easy to get lost in the terminology. Two names that often get mixed up are AG-UI and A2UI. They sound similar, but they actually serve very different purposes. Even better: they complement each other beautifully.CopilotKit has been working closely with Google on A2UI, and we’ll be shipping full support when the A2UI spec launches. But before that happens, let’s break down how these pieces fit into the broader agentic landscape.
Nathan Tarbert December 3, 2025
AWS Strands Agents Now Compatible with AG-UIThe agentic space has seen incredible momentum lately, but today’s update stands out: AWS Strands is now integrating with AG-UI and CopilotKit.
Uli Barkai and Nathan TarbertDecember 2, 2025
The State of Agentic UI: Comparing AG-UI, MCP-UI, and A2A ProtocolsOver 18 months ago, CopilotKit released its first version with one clear goal: give developers the tools to build intelligent, AI-native interfaces - not just bots with text boxes, but deeply integrated, usable, agentic UIs.
Nathan Tarbert November 20, 2025
How Microsoft Agent Framework + AG-UI Enable Agentic UX & Generative UIMicrosoft’s Agent Framework (MAF) is an open source agent framework that has just come on the scene, and offers exceptional ways to build agents and multi-step workflows in .NET or Python. AG-UI is a frontend/runtime layer that lets those agents show up in your app with a clean UI, streaming responses, and shared state, etc. The integration between the two is simple but powerful: MAF handles the reasoning and tool-use, and AG-UI bridges the interactions between the agent and the users.
Uli Barkai and Nathan TarbertNovember 11, 2025
Microsoft Agent Framework is now AG-UI Compatible!We’re excited to announce that Microsoft’s Agent Framework is now fully compatible with the AG-UI protocol - the open standard powering frontend-agent communication across the ecosystem. This integration means you can now connect Microsoft’s powerful reasoning and orchestration layer directly to rich, real-time user interfaces - without custom socket code, polling loops, or one-off APIs.
Bonnie and Nathan TarbertNovember 11, 2025
Build a Frontend for your Microsoft Agent Framework Agents with AG-UIIn this guide, you will learn how to build a frontend for your Microsoft Agent Framework Agents using AG-UI Protocol and CopilotKit. Microsoft Agent Framework will power the AI agents backend, while CopilotKit powers the frontend, and then AG-UI creates a bridge that enables the frontend to communicate with the backend.
Nathan Tarbert November 4, 2025
AG-UI Goes Mobile: The Kotlin SDK Unlocks Full Agent Connectivity Across Android, iOS, and JVMMark Fogle’s recent contribution of the Kotlin SDK to AG-UI is a big step forward: developers can now integrate AG-UI directly into Android, iOS, and JVM environments through a single, unified SDK.This means users can now connect to agents directly from mobile with no custom bridges, wrappers, or per-platform workarounds.
Anmol Baranwal and Nathan TarbertNovember 3, 2025
Master the 17 AG-UI Event Types for Building Agents the Right WayAgent-User Interaction Protocol (AG-UI) is quickly becoming the standard for connecting agents with user interfaces. It gives you a clean event stream to keep the UI in sync with what the agent is doing. All the communication is broken into typed events. I have been digging into the Protocol, especially around those core event types, to understand how everything fits together. Here’s what I picked up and why it matters.
Nathan Tarbert October 30, 2025
AG-UI Is Redefining the Agent–User Interaction LayerThe AG-UI Protocol continues to surge — and with it, a new standard is forming across the agentic landscape.AG-UI isn’t just another SDK. It’s the open protocol for agent-user communication, enabling real-time collaboration between humans and agents across any app, platform, or framework. What started as a developer-driven initiative inside the CopilotKit ecosystem has grown into an open-source movement — now being adopted, extended, and accelerated by leaders across the industry.
Bonnie and Eli BermanOctober 23, 2025
Building a Full-Stack AI Shopping Assistant with CrewAI and TavilyIn this guide, you will learn how to build a full-stack AI shopping assistant using CrewAI Flows, paired with the Tavily Search API in the backend. We'll then walk through adding a frontend to the shopping agent using CopilotKit to interact with it. The AI shopping assistant will be able to search products on commerce platforms such as Amazon, Target, or eBay- extract structured product data and provide tailored product recommendations.
Bonnie and Nathan TarbertOctober 3, 2025
Build a Stock Portfolio AI Agent (Fullstack, Pydantic AI + AG-UI)In this guide, you will learn how to build fullstack AI agents with Pydantic AI with the AG-UI protocol. Also, we will cover how to integrate the AG-UI + Pydantic AI agents with CopilotKit to chat with the agent and stream its responses in the frontend.
Bonnie and Nathan TarbertSeptember 25, 2025
Build a Frontend for your ADK Agents with AG-UIn this guide, you will learn how to build a frontend for your Agent Development Kit (ADK) Agents using AG-UI Protocol and CopilotKit. ADK will power the AI agents backend, while CopilotKit powers the frontend, and then AG-UI creates a bridge that enables the frontend to communicate with the backend.
Anmol Baranwal and Nathan TarbertSeptember 23, 2025
Here's How To Build Fullstack Agent Apps (Gemini, CopilotKit & LangGraph)AI agents are getting close to real world applications, but most developers still find it complex to build one. So we are going to build two practical agents: Post Generator that drafts LinkedIn/X content using live web search & Stack Analyzer that inspects GitHub repos and creates structured reports. We will be using Next.js frontend, FastAPI backend, CopilotKit, LangGraph workflows, and Google Gemini. You will find architecture, concepts, prompts, and practical stuff.
Bonnie and Nathan TarbertSeptember 22, 2025
Building a RAG-powered E-commerce Platform (LangGraph + MongoDB + CopilotKit)In this guide, you will learn how to build a RAG(Retrieval-Augmented Generation)-powered e-commerce platform using LangGraph, MongoDB Vector Search, and CopilotKit. The e-commerce platform helps users find products and get tailored product recommendations using LLMs.
Nathan Tarbert September 18, 2025
Building Fullstack Agentic Apps with Realtime Data: CopilotKit + TavilyToday, we are announcing the CopilotKit+Tavily partnership. Let's dive into why this makes so much sense.
Nathan Tarbert September 18, 2025
How CopilotKit + Mastra Enable Real-Time Agent InteractionToday, we are pleased to announce our partnership: CopilotKit 🤝 Mastra.
Eli Berman September 12, 2025
The Three Types of Generative UI: Static, Declarative and Fully GeneratedA practical breakdown for building reliable AI‑driven interfaces: when to use each, examples, and reality checks. Generative UI has been hyped as the future of how we’ll build software interfaces, but most of the conversation is still fuzzy. Should you let a model write your entire UI? Or lock it down with strict components? In reality, there are three clear approaches that fall along a spectrum of flexibility vs. control. This blog unpacks each one with examples, pros and cons, and some practical guidance on when to use which.
Nathan Tarbert and Eli BermanSeptember 10, 2025
Introducing the AG-UI Dojo: A Learning Tool to Help You Ship FasterBuilding AI agent frontends? You've probably hit the usual roadblocks-streaming updates that break, state bugs that make no sense, tool calls that require way too much setup. The AG-UI Dojo fixes that problem. It's a collection of working examples you can actually visualize and use.
Nathan Tarbert September 4, 2025
Introducing the AG-UI Protocol CLIBuilding agentic applications just got a whole lot easier. We’re excited to release the AG-UI protocol CLI flow- the quickest and easiest way to get started with AG-UI in under two minutes.
Itay Shemer and Eli BermanAugust 22, 2025
Case Study: Improving the “Vibecodability” of CopilotKitOutcomes: Faster, More Accurate Integrations • 66% faster integrations
The CopilotKit Expert MCP enabled a working, streaming AI agent frontend in a single shot, cutting LLM calls by two-thirds. We tested each approach by asking Cursor to create a CopilotKit UI, integrate it to a backend, and connect it to other components in an example application.
Bonnie and Nathan TarbertAugust 22, 2025
Build Your Own AI Stock Portfolio Agent with Agno + AG-UIIn this guide, you will learn how to integrate Agno agents with the AG-UI protocol. Additionally, we will cover how to integrate the AG-UI and Agno agents with CopilotKit, allowing users to chat with the agent and stream its responses in the frontend.
Eli Berman and Nathan TarbertAugust 21, 2025
Introducing the Vibe Coding Server + CopilotKit MCP ClientMaking it easier than ever to build agent-ready frontends (and vibe code with your coding agents). We’re excited to share a major update in CopilotKit’s MCP ecosystem: We’ve rebuilt the Open MCP Client and added a new Vibe Coding Server (optimized by Tadata). With these releases, developers can both connect their copilots to MCP servers seamlessly and give coding agents structured, reliable access to CopilotKit docs and examples, eliminating boilerplate, hallucinations, and friction.
Bonnie and Nathan TarbertAugust 13, 2025
Build a Fullstack Stock Portfolio Agent with LlamaIndex and AG-UIIn this guide, you will learn how to integrate LlamaIndex agents with the AG-UI protocol. Also, we will cover how to integrate the AG-UI + LlamaIndex agents with CopilotKit in order to chat with the agent and stream its responses to the frontend.
Bonnie and Nathan TarbertAugust 8, 2025
Build a Fullstack Stock Portfolio Agent with CrewAI and AG-UIIn this guide, you will learn how to integrate CrewAI agents with AG-UI protocol. Also, we will cover how to integrate the AG-UI + CrewAI agents with CopilotKit in order to chat with the agent and stream its responses in the frontend.
Eli Berman and Nathan TarbertAugust 7, 2025
Intro to AI AgentsA useful guide for those navigating the AI agent ecosystem. The Agent ecosystem can feel like alphabet soup (RAG, MCP, A2A, AG-UI, LangGraph, Tavily…). We've put together this guide for those trying to get into the ecosystem. 
Anmol Baranwal August 4, 2025
The Best AI Agent Resources You Should Know in 2025Most AI Agents used to break the moment you tried plugging them into a real app. The issue was the lack of shared protocols. But that’s changing fast. Over the past year, protocols like AG-UI (user interaction), A2A (agent-to-agent communication), and MCP (for tool access) have changed the ecosystem. So I took the chance to learn these protocols and understand how everything fits together. This post covers everything I picked up about protocols (AG-UI, ACP, A2A, MCP). You will also find a collection of the best educational repos for learning & building AI Agents. This list will give you the foundations and tools you need.
Nathan Tarbert August 4, 2025
What I Learned Building with AI Agents: 10 Hard-Earned Lessons for DevelopersBuilding AI-driven agents has been a trial by fire. As a developer, I’ve faced busted demos, odd bugs, and “oh no” moments that taught me more than any success did. That’s why I wanted to put together a list of 10 hard-earned lessons from the trenches of developing AI agents – hopefully, they save you some headaches.
Bonnie and Nathan TarbertAugust 1, 2025
Build a Fullstack Stock Portfolio Agent with LangGraph and AG-UIIn this guide, you will learn how to integrate a LangGraph agent with the AG-UI protocol. Also, we will cover how to integrate the AG-UI + LangGraph agent with CopilotKit in order to chat with the agent and stream its responses in the frontend.
Bonnie and Nathan TarbertJuly 30, 2025
Build a Full-Stack Stock Portfolio Agent with Mastra and AG-UIIn this guide, you will learn how to integrate a Mastra AI agent with the AG-UI protocol. Also, we will cover how to integrate the AG-UI + Mastra AI agent with CopilotKit in order to chat with the agent and stream its responses in the frontend.
Steven Hartland and Nathan TarbertJuly 23, 2025
Introducing Pydantic AI Integration with AG-UIWe’re excited to share that AG‑UI, the open Agent–User Interaction Protocol from CopilotKit, is now natively supported in **Pydantic AI,** thanks to a new integration built by Steven Hartland at Rocket Science, working hand in hand with the Pydantic AI team.
Ran Shem Tov June 19, 2025
Agent Streams Are Tricky, Here’s How We Got Ours to Make SenseOver the past year, we’ve spent a lot of time wiring agent frameworks into real frontends. stitching tool calls, tracking partial updates, and guessing when messages end. At some point, we realized we couldn’t keep patching forever.Here’s why, and how we started standardizing instead.
Laurie Voss June 18, 2025
Introducing the LlamaIndex Integration with AG-UI & CopilotKitIt’s been a big month for CopilotKit’s AG-UI project, which launched just over a month ago!
Since then, both CopilotKit and AG-UI have seen strong adoption, and today we’re celebrating with a wave of new features and updates. We’re also excited to announce that LlamaIndex is joining the celebration with the launch of their official AG-UI integration!
Ashpreet Bedi June 18, 2025
Introducing the Agno Integration with AG-UI & CopilotKitWe’re excited to announce the official CopilotKit integration with Agno via AG-UI — the open protocol for Agent-User Interactions. With this integration, you can now expose your Agno agents as AG-UI compatible apps.
Nathan Tarbert June 2, 2025
AG-UI Protocol: Bridging Agents to Any Front EndIn the past two years, powerful backend agents have entered the market. Tools like GitHub Copilot, Cursor, and Notion AI are already helping professionals write code faster, brainstorm content, and automate work. These are user-facing AI agents that collaborate with people in real time.
Uli Barkai June 2, 2025
Big Updates on AG-UI: Massive Traction and AdoptionWe launched AG-UI one month ago, and it’s been incredible to see the reception and adoption. Millions of views for content on AG-UI, tens of thousands of developers building, and new contributions & integrations in the pipeline by some of the most important players in the space, including **Agno, AWS, LlamaIndex, Alibaba, and more!**
Bonnie and Nathan TarbertMay 29, 2025
How to add a Frontend to any LangGraph Agent using AG-UI ProtocolIn this article, you will learn how to add a frontend to any LangGraph agent using the Agent User Interaction Protocol (AG-UI) developed by CopilotKit.
Bonnie and Nathan TarbertMay 29, 2025
How to add a Frontend to any CrewAI Agent using AG-UI ProtocolIn this article, you will learn how to add a frontend to any CrewAI agents using the Agent User Interaction Protocol (AG-UI) developed by CopilotKit.
Bonnie and Nathan TarbertMay 29, 2025
How to add a Frontend to any Mastra Agent using AG-UI ProtocolIn this article, you will learn how to add a frontend to any Mastra agent using the Agent User Interaction Protocol (AG-UI) developed by CopilotKit.
Bonnie and Nathan TarbertMay 29, 2025
How to add a Frontend to any AG2 Agent using AG-UI ProtocolIn this article, you will learn how to add a frontend to any AG2 agent using the Agent User Interaction Protocol (AG-UI) developed by CopilotKit.
Harish Mohan Raj May 12, 2025
Introducing AG2's Integration with CopilotKitWe’re excited to announce that CopilotKit now integrates with AG2, bringing together AG2’s multi-agent orchestration capabilities with CopilotKit’s React UI components. This integration creates a more seamless development experience for building AI-powered applications.
Nathan Tarbert May 12, 2025
Introducing AG-UI: The Protocol Where Agents Meet UsersToday, we are excited to announce the launch of our new open-protocol, AG-UI, the Agent-User Interaction Protocol.
A bridge between backend AI agents and full-stack applications. Where agents meet users.
Shane Thomas May 12, 2025
Introducing Mastra's Integration with CopilotKitIntroducing Mastra's integration with CopilotKit to bridge the backend agent with the user interactive layer.
Anmol Baranwal May 1, 2025
Add an MCP Client to Any React App in Under 30 MinutesAI Agents have unlocked powerful use cases so developers are automating complex workflows. But sometimes you just want to build smart stuff without all the extra complexity, which can be achieved using agentless architecture. Today, we will learn about what it means, how MCP fits in the whole scenario, and how to build our own agentless applications using CopilotKit (framework) with Composio servers.
Graham McBain April 8, 2025
🚀 Why CopilotKit is Going All-In on User-Interactive AI AgentsToday, we’re excited to share why CopilotKit is doubling down on user-interactive AI agents—copilots designed specifically to work alongside users rather than autonomously.
Graham McBain March 27, 2025
CopilotKit Expands Capabilities with CrewAI Integration for CoAgentsWe’re excited to share that you can now build a UI for all your Crews using CopilotKit.
Bonnie and Nathan TarbertMarch 27, 2025
How To Build Full-Stack AI Agents 🚀 (CrewAI + CopilotKit)In this article, you will learn how to build a full-stack restaurant finder AI agent that combines human-in-the-loop capabilities using CrewAI, CopilotKit, and Serper.
Bonnie and Nathan TarbertMarch 20, 2025
Easily Build a UI for Your AI Agent in Minutes (LangGraph + CopilotKit)In this article, you will learn how to build an agent-native research canvas app that combines human-in-the-loop capabilities using Langraph, CopilotKit, and Tavily.
Nathan Tarbert March 3, 2025
Build a RAG Copilot with MongoDB Vector Search & CopilotKitIn this tutorial, we’ll walk you through the process of building an AI-powered RAG Copilot for your product knowledge base using OpenAI API, MongoDB Atlas Vector Search, and CopilotKit.
Anmol Baranwal February 20, 2025
Agents 101: How to build your first AI Agent in 30 minutes!Building your own agent can be complex, and all the concepts, frameworks, and practices you need to follow can be a bit overwhelming. The good news is, that adding agents to your application can be easy and I'll show you how. In this guide, you will learn how to build your first AI agent in just 30 minutes, even if you don't know anything about AI agents.
Graham McBain February 18, 2025
Coagents V0.5: The Building Blocks of a Full-Stack AgentWe launched the CoAgents beta a few months ago, and it was a huge success right out of the gate. We've been working with Fortune 100 companies, thousands of individual developers, small teams, and a bunch of design partners. Through all those conversations and real-world use cases, we've identified the core building blocks for building a full-stack agent with LangGraph using CopilotKit.
Nathan Tarbert January 17, 2025
Build Your Own Knowledge-Based RAG CopilotIn this tutorial, we’ll walk you through the process of building an AI-powered Copilot for your product knowledge base using Anthropic AI API, Pinecone API, and CopilotKit.
Graham McBain December 17, 2024
The New Wave of Agent-Native AppsWe are witnessing the beginning of the most significant transformation in software since cloud and mobile: Agent-Native Applications (ANAs). This shift will fundamentally reshape the SaaS landscape even more rapidly than previous computing revolutions. Tools like LangChain Platform are making it easier than ever to build this new and exciting type of application.
Graham McBain December 10, 2024
CoAgents v0.2 release: Build Full-stack Agent Applications with LangGraph & CopilotKitAgentic Copilots are changing the SaaS landscape faster than any other previous computing revolution. With the latest release from CopilotKit with support for LangGraph JS and LangGraph Platform, it's easier than ever to build human-in-the-loop Agentic Copilots that are production-ready. These powerful tools let any company build the same experiences built by industry leaders like Cursor, Replit, and Github.
Graham McBain November 22, 2024
The Agentic Application Stack ExplainedThe Agentic Application Stack enables AI agents to power applications with autonomy and interactivity, detailing key layers like model serving, storage, tools, frameworks, hosting, and user experience.
Nathan Tarbert October 30, 2024
Build a Perplexity Clone with CopilotKit, Tavily & LangGraphIntegrate Agent-Native architecture into your app using CopilotKit, Tavily, and LangGraph. 
Atai Barkai October 22, 2024
Everything You Need To Build Agent-Native Applications (with LangGraph and CoAgents)CoAgents, the building blocks for the next generation of AI-native apps.
Atai Barkai September 25, 2024
Buildtime and Runtime Human-in-the-Loop AI (HITL)How HITL AI systems are bringing AI from demos to production.
Atai Barkai September 10, 2024
CoAgents: Connecting AI Agents to Realtime Application ContextDiscover how CoAgents, in partnership with LangGraph and CopilotKit, are revolutionizing the way AI agents interact within applications by aligning more closely with human input. This blog explores the essential role of human-in-the-loop systems to ensure AI agents perform optimally when they complement human efforts, rather than operating fully autonomously. Key Highlights: Human-AI Collaboration: Understand the significance of connecting AI agents with real-time application contexts to enhance decision-making and task performance. Monitoring and Steering: Learn how end-users can monitor and guide AI agents, ensuring they remain on the right track during operations.
Technology Integration: Dive into how integrating CoAgents with LangGraph elevates in-app human-in-the-loop agents, making advanced AI more accessible and effective for developers through CopilotKit.
Interactive Agent Features: Explore features such as streaming intermediate agent states for live updates, shared states for collaborative input, and agent-led interactive Q&As to refine operational responses.
Atai Barkai August 26, 2024
August’s Feature Rollout: Enhanced CopilotKit Capabilities Await! 😎Stay up to date with our latest monthly release! Discover the newest features, enhancements, and improvements we've rolled out in August to help you get the most out of our product. Dive into the details and see how these updates can benefit your experience.
Atai Barkai July 8, 2024
CopilotKit v1.0 Launch 🚀 GenUI, GraphQL protocol, React SDK hooksIntroducing CopilotKit V1, includes: Generative UI, GraphQL platform enhancements, React SDK hooks & more!
Atai Barkai July 8, 2024
🚀 Ceiling is being raised: analyzing our "superhuman code" that broke twitter (and why it's actually good code!)Explore the transformative impact of AI-enhanced coding with Atai Barkai's blog, which delves into the potent code developed by CopilotKit engineers that stirred considerable buzz on Twitter.
Richard Aragon and Atai BarkaiFebruary 28, 2024
🤩 THIS IS A PICTURE OF A PENGUIN: How GPT4, Gemini & LLaVA handle multimodal conflictExplore the intricate workings of multimodal AI systems in the detailed examination provided by the latest article on multimodal models. 
Atai Barkai December 5, 2023
🌐 RAG vs. Context-Window in GPT-4: accuracy, cost, & latencyExplore the advanced capabilities of LLMs in the latest analytical post that delves into the effectiveness of Retrieval Augmented Generation (RAG) versus GPT-4's context-window stuffing. Are you ready?
Stay in the know
Subscribe to our blog and get updates on CopilotKit in your inbox.
