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 ag

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 access and an authenticated environment 2. Language runtime : Node.js ^20.19.0 or =22.12.0, Python 3.11+, Go 1.24+, or a .NET Standard 2.0 compatible implementation 3. Go : GitHub Copilot CLI installed and authenticated ([Installation guide](https://docs.github.com/en/copilot/how tos/set up/install copilot cli)) The TypeScript, Python, and .NET packages use a bundled Copilot runtime by default, so they do not need a separate CLI installation. Installation Node.js/TypeScript Python Published Python wheels include a pinned runtime version. The pre download command caches that runtime locally; if skipped, the SDK attempts to download it automatically on first use. Go .NET Quick Start TypeScript Run: npx tsx index.ts Python Go .NET (C ) Run: dotnet run Streaming Responses Enable real time output for better UX: TypeScript Python Go .NET 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) Python (Pydantic) Go .NET (Microsoft.Extensions.AI) How Tools Work When Copilot decides to call your tool: 1. Copilot sends a tool call request with the parameters 2. The SDK runs your handler function 3. The result is sent back to Copilot 4. 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 Python 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 Python Go .NET Custom Agents Define specialized AI personas for specific tasks: TypeScript Python System Message Customize the AI's behavior and personality: TypeScript Python 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 Connect SDK to External Server TypeScript Python Go .NET Note: When configured to use an external server, the SDK manages only its connection and does not manage the external process. 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 Resume Session List and Delete Sessions Error Handling Graceful Shutdown Common Patterns Multi turn Conversation File Attachments Abort Long Operations Available Models Query available models at runtime: Best Practices 1. Always clean up : Use language native context managers or disposal, or explicitly disconnect sessions and stop clients 2. Set timeouts : Use sendAndWait with timeout for long operations 3. Handle events : Subscribe to error events for robust error handling 4. Use streaming : Enable streaming for better UX on long responses 5. Persist sessions : Use custom session IDs for multi turn conversations 6. Define clear tools : Write descriptive tool names and descriptions Architecture 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.

copilot-sdk