In this tutorial, we’ll explain how to use the Codename Goose AI Framework on Windows.
Codename Goose is an open-source AI agent that runs locally on your machine, automating tasks and integrating seamlessly with existing tools to boost productivity, streamline workflows, and enhance your development environment.
What is Codename Goose?
Codename Goose is an open-source, on-device AI agent designed to automate tasks and integrate seamlessly with a variety of tools and platforms. Its extensible architecture allows it to connect with numerous applications, enhancing productivity across different workflows.
Goose supports integration with several tools through its built-in and configurable extensions such JetBrains IDEs, Google Drive, Developer Tools, Developer Tools, Memory etc.
Additionally, you can use Goose to configure with various Large Language Model (LLM) providers, such as OpenAI’s GPT-4o, Gemini and Anthropic’s Claude Sonnet, and integrate their preferred models.
With Goose’s modular design and support for the Model Context Protocol (MCP), you can connect with a wide ecosystem of capabilities, making it a versatile tool for automating complex workflows.
How Codename Goose Works?
As Goose operates an intelligent agent that capable of handling complex tasks through coordination using it’s core features such as Model Context Protocol, LLM Providers and Extensions:
- Model Context Protocol (MCP): This is an open standard for interoperability between AI agents and external tools. MCP servers expose functions or data that Goose can invoke as per requirement.
- LLM Providers: Goose is compatible with wide range of LLM provides as it’s supports cloud and local providers.
- Extensions: The extensions are key to Goose’s adaptability that provides the ability to connect with applications and tools we already use. You can connect with GitHub, access Google Drive or integrating with JetBrains IDEs with extensive.
How to Install Codename Goose on Windows?
AS there are no nativ installation support for windows, however we can install and use Goose on Windows using WSL (Windows Subsystem for Linux). Here are steps to install Goose on Windos using WSL:
.
1. Install WSL
You need to open command line as Administrator and install WSL and the default Ubuntu distribution using following command:
wsl --install
Need to restart your computer if prompted.
2. Run Goose Installation
After installing WSL, you need to run Goose script using below to install it.
curl -fsSL https://github.com/block/goose/releases/download/stable/download_cli.sh | bash
3. Setup LLM Provider
After sucessfull installation, you need to setup LLM providers towork Goose with a set of supported LLM providers. You’ll need an API key to get started. When you use Goose for the first time, you’ll be prompted to select a provider and enter your API key.
Upon installing, the Provider screen will appear. Here is where you can choose your LLM Provider in Goose Desktop.
After selecting your provider, you’ll be prompted to enter an API key if applicable. Do so, and click Submit button.
4. Running Goose
In Goose Desktop, starting a session is straightforward. After choosing your provider, you’ll see the session interface ready for use.
You can type your questions, tasks, or instructions directly into the input field, and Goose will get to work immediately.
Final Words
Here in this tutorial, we have explained about wht is Goose and how to install it on Windows enviroment to get started with it. You can also configure Extensions to extend Goose’s functionality, including adding new ones or toggling them on and off.