Copilot SDK

Build AI Copilots for Your Product

Production-ready AI Copilots for any product. Connect any LLM, deploy on your infrastructure, own your data. Built for speed and control.

Documentation

Quick Start

Install

npm install @yourgpt/copilot-sdk @yourgpt/llm-sdk openai

Frontend (React)

import { CopilotProvider } from "@yourgpt/copilot-sdk/react";
import { CopilotChat } from "@yourgpt/copilot-sdk/ui";
import "@yourgpt/copilot-sdk/ui/styles.css";

function App() {
  return (
    <CopilotProvider runtimeUrl="/api/chat">
      <YourApp />
      <CopilotChat />
    </CopilotProvider>
  );
}

Backend (Next.js)

// app/api/chat/route.ts
import { streamText } from "@yourgpt/llm-sdk";
import { openai } from "@yourgpt/llm-sdk/openai";

export async function POST(req: Request) {
  const { messages } = await req.json();

  const result = await streamText({
    model: openai("gpt-4o"),
    system: "You are a helpful assistant.",
    messages,
  });

  return result.toTextStreamResponse();
}

LLM SDK - Multi-Provider Support

Use any LLM provider with a unified API:

import { streamText } from "@yourgpt/llm-sdk";
import { openai } from "@yourgpt/llm-sdk/openai";
import { anthropic } from "@yourgpt/llm-sdk/anthropic";
import { google } from "@yourgpt/llm-sdk/google";
import { xai } from "@yourgpt/llm-sdk/xai";

// OpenAI
await streamText({ model: openai("gpt-4o"), messages });

// Anthropic
await streamText({ model: anthropic("claude-sonnet-4-20250514"), messages });

// Google Gemini (uses OpenAI-compatible API)
await streamText({ model: google("gemini-2.0-flash"), messages });

// xAI Grok (uses OpenAI-compatible API)
await streamText({ model: xai("grok-3-fast-beta"), messages });

SDK Requirements

Provider	SDK Required
OpenAI, Google, xAI	openai
Anthropic	@anthropic-ai/sdk

Server-Side Tools

Add tools to let the AI call functions on your server:

import { streamText, tool } from "@yourgpt/llm-sdk";
import { openai } from "@yourgpt/llm-sdk/openai";
import { z } from "zod";

const result = await streamText({
  model: openai("gpt-4o"),
  messages,
  tools: {
    getWeather: tool({
      description: "Get current weather for a city",
      parameters: z.object({
        city: z.string().describe("City name"),
      }),
      execute: async ({ city }) => {
        const data = await fetchWeatherAPI(city);
        return { temperature: data.temp, condition: data.condition };
      },
    }),
  },
  maxSteps: 5,
});

return result.toDataStreamResponse();

Packages

Package	Description
@yourgpt/copilot-sdk/react	React hooks and provider
@yourgpt/copilot-sdk/ui	Pre-built chat components
@yourgpt/copilot-sdk/core	Core utilities and context capture tools
@yourgpt/llm-sdk	Multi-provider LLM SDK with streaming

Documentation

Visit copilot-sdk.yourgpt.ai for full documentation.

• Getting Started
• Why Copilot SDK
• LLM SDK
• Providers
• API Reference

Contributing

Have any feedback? Share it with us.

@0fficialRohit · @rege_dev

License

MIT License