copilotkit-debug

Enables rendering of A2UI (Agent-to-UI) declarative surfaces in CopilotKit v2, covering runtime configuration, theme customization, surface deduplication, and action-bridge cleanup patterns.

A framework for building agent-native applications with generative UI, shared state between human and AI, and human-in-the-loop workflows, providing React components for AI copilot integration.

Guide for building AI-powered features with CopilotKit v2 — including chat interfaces, frontend tool registration, sharing application context with agents, handling interrupts, and runtime configuration.

Reference for @copilotkit/runtime — mounting a fetch-native CopilotRuntime on any JS server, wiring middleware, configuring agent runners, and integrating 12+ external agent frameworks including LangGraph, CrewAI, and Mastra.

Covers wiring external agent frameworks (LangGraph, CrewAI, PydanticAI, Mastra, ADK, LlamaIndex, Agno, and others) into CopilotKit applications via the AG-UI streaming protocol.

A CopilotKit skill for upgrading between SDK versions, part of the framework behind the AG-UI Protocol adopted by Google, LangChain, and AWS for full-stack agentic applications.

Guides contributors through the CopilotKit open-source monorepo workflow — forking, setting up the pnpm workspace, building packages, running Vitest tests, and submitting PRs with conventional commits. Includes an MCP server for live documentation and source code search.

Teaches the AG-UI (Agent-User Interaction) protocol for CopilotKit, covering event-based agent-to-UI communication via SSE, state synchronization, tool calls, and human-in-the-loop flows.

Step-by-step guide for adding CopilotKit to existing projects or bootstrapping new ones — covers framework detection, package installation, runtime wiring, provider setup, and first working chat integration.

A CopilotKit skill for keeping the SDK up to date, part of the framework for building agent-native applications with generative UI, shared state, and human-in-the-loop workflows.