PaperclipOrg
Claude Agent Skill · by Github

Copilot Sdk

If you need to embed Copilot's agent runtime in your own application, this skill walks you through the GitHub Copilot SDK across Python, TypeScript, Go, and .NE

Install
Terminal · npx
$npx skills add https://github.com/github/awesome-copilot --skill copilot-sdk
Works with Paperclip

How Copilot Sdk fits into a Paperclip company.

Copilot Sdk drops into any Paperclip agent that handles this kind of work. Assign it to a specialist inside a pre-configured PaperclipOrg company and the skill becomes available on every heartbeat — no prompt engineering, no tool wiring.

S
SaaS FactoryPaired

Pre-configured AI company — 18 agents, 18 skills, one-time purchase.

$27$59
Explore pack
Source file
SKILL.md914 lines
Expand
---name: copilot-sdkdescription: Build agentic applications with GitHub Copilot SDK. Use when embedding AI agents in apps, creating custom tools, implementing streaming responses, managing sessions, connecting to MCP servers, or creating custom agents. Triggers on Copilot SDK, GitHub SDK, agentic app, embed Copilot, programmable agent, MCP server, custom agent.--- # GitHub Copilot SDK Embed Copilot's agentic workflows in any application using Python, TypeScript, Go, or .NET. ## Overview The GitHub Copilot SDK exposes the same engine behind Copilot CLI: a production-tested agent runtime you can invoke programmatically. No need to build your own orchestration - you define agent behavior, Copilot handles planning, tool invocation, file edits, and more. ## Prerequisites 1. **GitHub Copilot CLI** installed and authenticated ([Installation guide](https://docs.github.com/en/copilot/how-tos/set-up/install-copilot-cli))2. **Language runtime**: Node.js 18+, Python 3.8+, Go 1.21+, or .NET 8.0+ Verify CLI: `copilot --version` ## Installation ### Node.js/TypeScript```bashmkdir copilot-demo && cd copilot-demonpm init -y --init-type modulenpm install @github/copilot-sdk tsx``` ### Python```bashpip install github-copilot-sdk``` ### Go```bashmkdir copilot-demo && cd copilot-demogo mod init copilot-demogo get github.com/github/copilot-sdk/go``` ### .NET```bashdotnet new console -n CopilotDemo && cd CopilotDemodotnet add package GitHub.Copilot.SDK``` ## Quick Start ### TypeScript```typescriptimport { CopilotClient, approveAll } from "@github/copilot-sdk"; const client = new CopilotClient();const session = await client.createSession({    onPermissionRequest: approveAll,    model: "gpt-4.1",}); const response = await session.sendAndWait({ prompt: "What is 2 + 2?" });console.log(response?.data.content); await client.stop();process.exit(0);``` Run: `npx tsx index.ts` ### Python```pythonimport asynciofrom copilot import CopilotClient, PermissionHandler async def main():    client = CopilotClient()    await client.start()     session = await client.create_session({        "on_permission_request": PermissionHandler.approve_all,        "model": "gpt-4.1",    })    response = await session.send_and_wait({"prompt": "What is 2 + 2?"})     print(response.data.content)    await client.stop() asyncio.run(main())``` ### Go```gopackage main import (    "fmt"    "log"    "os"    copilot "github.com/github/copilot-sdk/go") func main() {    client := copilot.NewClient(nil)    if err := client.Start(); err != nil {        log.Fatal(err)    }    defer client.Stop()     session, err := client.CreateSession(&copilot.SessionConfig{        OnPermissionRequest: copilot.PermissionHandler.ApproveAll,        Model:               "gpt-4.1",    })    if err != nil {        log.Fatal(err)    }     response, err := session.SendAndWait(copilot.MessageOptions{Prompt: "What is 2 + 2?"}, 0)    if err != nil {        log.Fatal(err)    }     fmt.Println(*response.Data.Content)    os.Exit(0)}``` ### .NET (C#)```csharpusing GitHub.Copilot.SDK; await using var client = new CopilotClient();await using var session = await client.CreateSessionAsync(new SessionConfig{    OnPermissionRequest = PermissionHandler.ApproveAll,    Model = "gpt-4.1",}); var response = await session.SendAndWaitAsync(new MessageOptions { Prompt = "What is 2 + 2?" });Console.WriteLine(response?.Data.Content);``` Run: `dotnet run` ## Streaming Responses Enable real-time output for better UX: ### TypeScript```typescriptimport { CopilotClient, approveAll, SessionEvent } from "@github/copilot-sdk"; const client = new CopilotClient();const session = await client.createSession({    onPermissionRequest: approveAll,    model: "gpt-4.1",    streaming: true,}); session.on((event: SessionEvent) => {    if (event.type === "assistant.message_delta") {        process.stdout.write(event.data.deltaContent);    }    if (event.type === "session.idle") {        console.log(); // New line when done    }}); await session.sendAndWait({ prompt: "Tell me a short joke" }); await client.stop();process.exit(0);``` ### Python```pythonimport asyncioimport sysfrom copilot import CopilotClient, PermissionHandlerfrom copilot.generated.session_events import SessionEventType async def main():    client = CopilotClient()    await client.start()     session = await client.create_session({        "on_permission_request": PermissionHandler.approve_all,        "model": "gpt-4.1",        "streaming": True,    })     def handle_event(event):        if event.type == SessionEventType.ASSISTANT_MESSAGE_DELTA:            sys.stdout.write(event.data.delta_content)            sys.stdout.flush()        if event.type == SessionEventType.SESSION_IDLE:            print()     session.on(handle_event)    await session.send_and_wait({"prompt": "Tell me a short joke"})    await client.stop() asyncio.run(main())``` ### Go```gosession, err := client.CreateSession(&copilot.SessionConfig{	OnPermissionRequest: copilot.PermissionHandler.ApproveAll,    Model:     "gpt-4.1",    Streaming: true,}) session.On(func(event copilot.SessionEvent) {    if event.Type == "assistant.message_delta" {        fmt.Print(*event.Data.DeltaContent)    }    if event.Type == "session.idle" {        fmt.Println()    }}) _, err = session.SendAndWait(copilot.MessageOptions{Prompt: "Tell me a short joke"}, 0)``` ### .NET```csharpawait using var session = await client.CreateSessionAsync(new SessionConfig{    OnPermissionRequest = PermissionHandler.ApproveAll,    Model = "gpt-4.1",    Streaming = true,}); session.On(ev =>{    if (ev is AssistantMessageDeltaEvent deltaEvent)        Console.Write(deltaEvent.Data.DeltaContent);    if (ev is SessionIdleEvent)        Console.WriteLine();}); await session.SendAndWaitAsync(new MessageOptions { Prompt = "Tell me a short joke" });``` ## Custom Tools Define tools that Copilot can invoke during reasoning. When you define a tool, you tell Copilot:1. **What the tool does** (description)2. **What parameters it needs** (schema)3. **What code to run** (handler) ### TypeScript (JSON Schema)```typescriptimport { CopilotClient, approveAll, defineTool, SessionEvent } from "@github/copilot-sdk"; const getWeather = defineTool("get_weather", {    description: "Get the current weather for a city",    parameters: {        type: "object",        properties: {            city: { type: "string", description: "The city name" },        },        required: ["city"],    },    handler: async (args: { city: string }) => {        const { city } = args;        // In a real app, call a weather API here        const conditions = ["sunny", "cloudy", "rainy", "partly cloudy"];        const temp = Math.floor(Math.random() * 30) + 50;        const condition = conditions[Math.floor(Math.random() * conditions.length)];        return { city, temperature: `${temp}°F`, condition };    },}); const client = new CopilotClient();const session = await client.createSession({    onPermissionRequest: approveAll,    model: "gpt-4.1",    streaming: true,    tools: [getWeather],}); session.on((event: SessionEvent) => {    if (event.type === "assistant.message_delta") {        process.stdout.write(event.data.deltaContent);    }}); await session.sendAndWait({    prompt: "What's the weather like in Seattle and Tokyo?",}); await client.stop();process.exit(0);``` ### Python (Pydantic)```pythonimport asyncioimport randomimport sysfrom copilot import CopilotClient, PermissionHandlerfrom copilot.tools import define_toolfrom copilot.generated.session_events import SessionEventTypefrom pydantic import BaseModel, Field class GetWeatherParams(BaseModel):    city: str = Field(description="The name of the city to get weather for") @define_tool(description="Get the current weather for a city")async def get_weather(params: GetWeatherParams) -> dict:    city = params.city    conditions = ["sunny", "cloudy", "rainy", "partly cloudy"]    temp = random.randint(50, 80)    condition = random.choice(conditions)    return {"city": city, "temperature": f"{temp}°F", "condition": condition} async def main():    client = CopilotClient()    await client.start()     session = await client.create_session({        "on_permission_request": PermissionHandler.approve_all,        "model": "gpt-4.1",        "streaming": True,        "tools": [get_weather],    })     def handle_event(event):        if event.type == SessionEventType.ASSISTANT_MESSAGE_DELTA:            sys.stdout.write(event.data.delta_content)            sys.stdout.flush()     session.on(handle_event)     await session.send_and_wait({        "prompt": "What's the weather like in Seattle and Tokyo?"    })     await client.stop() asyncio.run(main())``` ### Go```gotype WeatherParams struct {    City string `json:"city" jsonschema:"The city name"`} type WeatherResult struct {    City        string `json:"city"`    Temperature string `json:"temperature"`    Condition   string `json:"condition"`} getWeather := copilot.DefineTool(    "get_weather",    "Get the current weather for a city",    func(params WeatherParams, inv copilot.ToolInvocation) (WeatherResult, error) {        conditions := []string{"sunny", "cloudy", "rainy", "partly cloudy"}        temp := rand.Intn(30) + 50        condition := conditions[rand.Intn(len(conditions))]        return WeatherResult{            City:        params.City,            Temperature: fmt.Sprintf("%d°F", temp),            Condition:   condition,        }, nil    },) session, _ := client.CreateSession(&copilot.SessionConfig{	OnPermissionRequest: copilot.PermissionHandler.ApproveAll,    Model:     "gpt-4.1",    Streaming: true,    Tools:     []copilot.Tool{getWeather},})``` ### .NET (Microsoft.Extensions.AI)```csharpusing GitHub.Copilot.SDK;using Microsoft.Extensions.AI;using System.ComponentModel; var getWeather = AIFunctionFactory.Create(    ([Description("The city name")] string city) =>    {        var conditions = new[] { "sunny", "cloudy", "rainy", "partly cloudy" };        var temp = Random.Shared.Next(50, 80);        var condition = conditions[Random.Shared.Next(conditions.Length)];        return new { city, temperature = $"{temp}°F", condition };    },    "get_weather",    "Get the current weather for a city"); await using var session = await client.CreateSessionAsync(new SessionConfig{    OnPermissionRequest = PermissionHandler.ApproveAll,    Model = "gpt-4.1",    Streaming = true,    Tools = [getWeather],});``` ## How Tools Work When Copilot decides to call your tool:1. Copilot sends a tool call request with the parameters2. The SDK runs your handler function3. The result is sent back to Copilot4. Copilot incorporates the result into its response Copilot decides when to call your tool based on the user's question and your tool's description. ## Interactive CLI Assistant Build a complete interactive assistant: ### TypeScript```typescriptimport { CopilotClient, approveAll, defineTool, SessionEvent } from "@github/copilot-sdk";import * as readline from "readline"; const getWeather = defineTool("get_weather", {    description: "Get the current weather for a city",    parameters: {        type: "object",        properties: {            city: { type: "string", description: "The city name" },        },        required: ["city"],    },    handler: async ({ city }) => {        const conditions = ["sunny", "cloudy", "rainy", "partly cloudy"];        const temp = Math.floor(Math.random() * 30) + 50;        const condition = conditions[Math.floor(Math.random() * conditions.length)];        return { city, temperature: `${temp}°F`, condition };    },}); const client = new CopilotClient();const session = await client.createSession({    onPermissionRequest: approveAll,    model: "gpt-4.1",    streaming: true,    tools: [getWeather],}); session.on((event: SessionEvent) => {    if (event.type === "assistant.message_delta") {        process.stdout.write(event.data.deltaContent);    }}); const rl = readline.createInterface({    input: process.stdin,    output: process.stdout,}); console.log("Weather Assistant (type 'exit' to quit)");console.log("Try: 'What's the weather in Paris?'\n"); const prompt = () => {    rl.question("You: ", async (input) => {        if (input.toLowerCase() === "exit") {            await client.stop();            rl.close();            return;        }         process.stdout.write("Assistant: ");        await session.sendAndWait({ prompt: input });        console.log("\n");        prompt();    });}; prompt();``` ### Python```pythonimport asyncioimport randomimport sysfrom copilot import CopilotClient, PermissionHandlerfrom copilot.tools import define_toolfrom copilot.generated.session_events import SessionEventTypefrom pydantic import BaseModel, Field class GetWeatherParams(BaseModel):    city: str = Field(description="The name of the city to get weather for") @define_tool(description="Get the current weather for a city")async def get_weather(params: GetWeatherParams) -> dict:    conditions = ["sunny", "cloudy", "rainy", "partly cloudy"]    temp = random.randint(50, 80)    condition = random.choice(conditions)    return {"city": params.city, "temperature": f"{temp}°F", "condition": condition} async def main():    client = CopilotClient()    await client.start()     session = await client.create_session({        "on_permission_request": PermissionHandler.approve_all,        "model": "gpt-4.1",        "streaming": True,        "tools": [get_weather],    })     def handle_event(event):        if event.type == SessionEventType.ASSISTANT_MESSAGE_DELTA:            sys.stdout.write(event.data.delta_content)            sys.stdout.flush()     session.on(handle_event)     print("Weather Assistant (type 'exit' to quit)")    print("Try: 'What's the weather in Paris?'\n")     while True:        try:            user_input = input("You: ")        except EOFError:            break         if user_input.lower() == "exit":            break         sys.stdout.write("Assistant: ")        await session.send_and_wait({"prompt": user_input})        print("\n")     await client.stop() asyncio.run(main())``` ## MCP Server Integration Connect to MCP (Model Context Protocol) servers for pre-built tools. Connect to GitHub's MCP server for repository, issue, and PR access: ### TypeScript```typescriptconst session = await client.createSession({    onPermissionRequest: approveAll,    model: "gpt-4.1",    mcpServers: {        github: {            type: "http",            url: "https://api.githubcopilot.com/mcp/",        },    },});``` ### Python```pythonsession = await client.create_session({    "on_permission_request": PermissionHandler.approve_all,    "model": "gpt-4.1",    "mcp_servers": {        "github": {            "type": "http",            "url": "https://api.githubcopilot.com/mcp/",        },    },})``` ### Go```gosession, _ := client.CreateSession(&copilot.SessionConfig{	OnPermissionRequest: copilot.PermissionHandler.ApproveAll,    Model: "gpt-4.1",    MCPServers: map[string]copilot.MCPServerConfig{        "github": {            "type": "http",            "url": "https://api.githubcopilot.com/mcp/",        },    },})``` ### .NET```csharpawait using var session = await client.CreateSessionAsync(new SessionConfig{    OnPermissionRequest = PermissionHandler.ApproveAll,    Model = "gpt-4.1",    McpServers = new Dictionary<string, McpServerConfig>    {        ["github"] = new McpServerConfig        {            Type = "http",            Url = "https://api.githubcopilot.com/mcp/",        },    },});``` ## Custom Agents Define specialized AI personas for specific tasks: ### TypeScript```typescriptconst session = await client.createSession({    onPermissionRequest: approveAll,    model: "gpt-4.1",    customAgents: [{        name: "pr-reviewer",        displayName: "PR Reviewer",        description: "Reviews pull requests for best practices",        prompt: "You are an expert code reviewer. Focus on security, performance, and maintainability.",    }],});``` ### Python```pythonsession = await client.create_session({    "on_permission_request": PermissionHandler.approve_all,    "model": "gpt-4.1",    "custom_agents": [{        "name": "pr-reviewer",        "display_name": "PR Reviewer",        "description": "Reviews pull requests for best practices",        "prompt": "You are an expert code reviewer. Focus on security, performance, and maintainability.",    }],})``` ## System Message Customize the AI's behavior and personality: ### TypeScript```typescriptconst session = await client.createSession({    onPermissionRequest: approveAll,    model: "gpt-4.1",    systemMessage: {        content: "You are a helpful assistant for our engineering team. Always be concise.",    },});``` ### Python```pythonsession = await client.create_session({    "on_permission_request": PermissionHandler.approve_all,    "model": "gpt-4.1",    "system_message": {        "content": "You are a helpful assistant for our engineering team. Always be concise.",    },})``` ## External CLI Server Run the CLI in server mode separately and connect the SDK to it. Useful for debugging, resource sharing, or custom environments. ### Start CLI in Server Mode```bashcopilot --server --port 4321``` ### Connect SDK to External Server #### TypeScript```typescriptconst client = new CopilotClient({    cliUrl: "localhost:4321"}); const session = await client.createSession({    onPermissionRequest: approveAll,    model: "gpt-4.1",});``` #### Python```pythonclient = CopilotClient({    "cli_url": "localhost:4321"})await client.start() session = await client.create_session({    "on_permission_request": PermissionHandler.approve_all,    "model": "gpt-4.1",})``` #### Go```goclient := copilot.NewClient(&copilot.ClientOptions{    CLIUrl: "localhost:4321",}) if err := client.Start(); err != nil {    log.Fatal(err)} session, _ := client.CreateSession(&copilot.SessionConfig{	OnPermissionRequest: copilot.PermissionHandler.ApproveAll,	Model:               "gpt-4.1",})``` #### .NET```csharpusing var client = new CopilotClient(new CopilotClientOptions{    CliUrl = "localhost:4321"}); await using var session = await client.CreateSessionAsync(new SessionConfig{    OnPermissionRequest = PermissionHandler.ApproveAll,    Model = "gpt-4.1",});``` **Note:** When `cliUrl` is provided, the SDK will not spawn or manage a CLI process - it only connects to the existing server. ## Event Types | Event | Description ||-------|-------------|| `user.message` | User input added || `assistant.message` | Complete model response || `assistant.message_delta` | Streaming response chunk || `assistant.reasoning` | Model reasoning (model-dependent) || `assistant.reasoning_delta` | Streaming reasoning chunk || `tool.execution_start` | Tool invocation started || `tool.execution_complete` | Tool execution finished || `session.idle` | No active processing || `session.error` | Error occurred | ## Client Configuration | Option | Description | Default ||--------|-------------|---------|| `cliPath` | Path to Copilot CLI executable | System PATH || `cliUrl` | Connect to existing server (e.g., "localhost:4321") | None || `port` | Server communication port | Random || `useStdio` | Use stdio transport instead of TCP | true || `logLevel` | Logging verbosity | "info" || `autoStart` | Launch server automatically | true || `autoRestart` | Restart on crashes | true || `cwd` | Working directory for CLI process | Inherited | ## Session Configuration | Option | Description ||--------|-------------|| `model` | LLM to use ("gpt-4.1", "claude-sonnet-4.5", etc.) || `sessionId` | Custom session identifier || `tools` | Custom tool definitions || `mcpServers` | MCP server connections || `customAgents` | Custom agent personas || `systemMessage` | Override default system prompt || `streaming` | Enable incremental response chunks || `availableTools` | Whitelist of permitted tools || `excludedTools` | Blacklist of disabled tools | ## Session Persistence Save and resume conversations across restarts: ### Create with Custom ID```typescriptconst session = await client.createSession({    onPermissionRequest: approveAll,    sessionId: "user-123-conversation",    model: "gpt-4.1"});``` ### Resume Session```typescriptconst session = await client.resumeSession("user-123-conversation", { onPermissionRequest: approveAll });await session.send({ prompt: "What did we discuss earlier?" });``` ### List and Delete Sessions```typescriptconst sessions = await client.listSessions();await client.deleteSession("old-session-id");``` ## Error Handling ```typescripttry {    const client = new CopilotClient();    const session = await client.createSession({        onPermissionRequest: approveAll,        model: "gpt-4.1",    });    const response = await session.sendAndWait(        { prompt: "Hello!" },        30000 // timeout in ms    );} catch (error) {    if (error.code === "ENOENT") {        console.error("Copilot CLI not installed");    } else if (error.code === "ECONNREFUSED") {        console.error("Cannot connect to Copilot server");    } else {        console.error("Error:", error.message);    }} finally {    await client.stop();}``` ## Graceful Shutdown ```typescriptprocess.on("SIGINT", async () => {    console.log("Shutting down...");    await client.stop();    process.exit(0);});``` ## Common Patterns ### Multi-turn Conversation```typescriptconst session = await client.createSession({    onPermissionRequest: approveAll,    model: "gpt-4.1",}); await session.sendAndWait({ prompt: "My name is Alice" });await session.sendAndWait({ prompt: "What's my name?" });// Response: "Your name is Alice"``` ### File Attachments```typescriptawait session.send({    prompt: "Analyze this file",    attachments: [{        type: "file",        path: "./data.csv",        displayName: "Sales Data"    }]});``` ### Abort Long Operations```typescriptconst timeoutId = setTimeout(() => {    session.abort();}, 60000); session.on((event) => {    if (event.type === "session.idle") {        clearTimeout(timeoutId);    }});``` ## Available Models Query available models at runtime: ```typescriptconst models = await client.getModels();// Returns: ["gpt-4.1", "gpt-4o", "claude-sonnet-4.5", ...]``` ## Best Practices 1. **Always cleanup**: Use `try-finally` or `defer` to ensure `client.stop()` is called2. **Set timeouts**: Use `sendAndWait` with timeout for long operations3. **Handle events**: Subscribe to error events for robust error handling4. **Use streaming**: Enable streaming for better UX on long responses5. **Persist sessions**: Use custom session IDs for multi-turn conversations6. **Define clear tools**: Write descriptive tool names and descriptions ## Architecture ```Your Application       |  SDK Client       | JSON-RPC  Copilot CLI (server mode)       |  GitHub (models, auth)``` The SDK manages the CLI process lifecycle automatically. All communication happens via JSON-RPC over stdio or TCP. ## Resources - **GitHub Repository**: https://github.com/github/copilot-sdk- **Getting Started Tutorial**: https://github.com/github/copilot-sdk/blob/main/docs/tutorials/first-app.md- **GitHub MCP Server**: https://github.com/github/github-mcp-server- **MCP Servers Directory**: https://github.com/modelcontextprotocol/servers- **Cookbook**: https://github.com/github/copilot-sdk/tree/main/cookbook- **Samples**: https://github.com/github/copilot-sdk/tree/main/samples ## Status This SDK is in **Technical Preview** and may have breaking changes. Not recommended for production use yet.