CopilotKit
CopilotKit is an SDK for building agent-native applications: the layer between your agents and your users. It started as a React library and now supports React, Angular, Vue and React Native, plus Slack and Microsoft Teams, so one agent can power a web app, a mobile app and a team workspace. The company behind it also created the AG-UI protocol for connecting agents to user interfaces.
It suits front-end and full-stack teams adding AI features to their own products rather than building a chatbot from scratch. Features include a customisable chat UI with streaming and tool calls, backend tools that return UI components rendered in the client, generative UI that agents update at runtime, a shared state layer that agents and components both read and write, and human-in-the-loop steps where the agent pauses for user input. It works with existing agent frameworks rather than replacing them.
Getting started: ask your coding agent to run `npx --yes copilotkit@latest onboard start` in your project, or scaffold a new project with `npx copilotkit@latest create`. Agent skills for coding assistants are available via `npx copilotkit@latest skills install`.
Features
Why choose it
Trade-offs
Maintenance check
- Last commit
- 2026-11-02
- Commits, last 90 days
- 100+
- Releases, last 90 days
- 30
- Latest release
- v1.83.0 (2026-10-28)
- Open issues
- 374
- Recent issues with at least one comment
- 8/8
Oldest week on the left; the latest four weeks are orange. Source: GitHub commit activity.
Status rule: archived on GitHub = Archived; last commit within 90 days = Active; 91 to 365 days = Slowing; older = Stale. All figures come from the GitHub API on the date shown.
Most discussed open issues
- #1881 🚀 Feature Request: Mastra Message History Support32 comments · 2025-05-27
- #6408 🐛 Bug: v1 surfaces orphaned by the v1.50.0 re-implementation (readable context; server-side actions + MCP)17 comments · 2026-08-05
Issue titles are quoted from GitHub. Open the link to check whether a problem still applies to the version you use.
Who it fits
- You are building a React, Angular, Vue, or React Native app and need an MIT-licensed frontend layer for in-app AI copilots, generative UI, and human-in-the-loop state sharing.
- You want to connect your existing LangGraph, CrewAI, or custom agent backend to web apps, Slack, and Microsoft Teams via the AG-UI protocol.
- You are looking for a backend agent orchestration engine rather than a frontend/UI integration SDK; use CrewAI, Pydantic AI, or LangGraph for the backend.
- You want a standalone desktop chat client for personal use; use Cherry Studio.
How it differs
vs CrewAI★ 60.4kCopilotKit is the frontend and user-interaction layer (Generative UI, shared state, AG-UI) that connects your app to an agent; CrewAI is the backend Python framework that orchestrates the agents themselves. They are often used together.
Compatibility
Quick start
Use cases
Alternatives
Comments
No comments yet. Be the first!