AI Assistants Dev Container Features

This repository provides dev container Features for installing AI coding assistants in your development containers.

Features

• Goose: Install the Goose AI coding assistant from Block in your dev container
• Gemini CLI: Install the Google Gemini CLI for AI-powered development

Usage

Using Goose

{
    "name": "My Dev Container",
    "image": "mcr.microsoft.com/devcontainers/base:ubuntu",
    "features": {
        "ghcr.io/mikegehard/ai-assistants-devcontainers-feature/goose:1.0": {}
    },
    "runArgs": ["--env-file",".env"]
}

Using Gemini CLI

{
    "name": "My Dev Container",
    "image": "mcr.microsoft.com/devcontainers/base:ubuntu",
    "features": {
        "ghcr.io/mikegehard/ai-assistants-devcontainers-feature/gemini-cli:1.0": {}
    },
    "runArgs": ["--env-file",".env"]
}

Using Both Features

{
    "name": "My Dev Container",
    "image": "mcr.microsoft.com/devcontainers/base:ubuntu",
    "features": {
        "ghcr.io/mikegehard/ai-assistants-devcontainers-feature/goose:1.0": {},
        "ghcr.io/mikegehard/ai-assistants-devcontainers-feature/gemini-cli:1.0": {}
    },
    "runArgs": ["--env-file",".env"]
}

API Keys

Goose API Keys

For Goose to function properly, you'll need to set up API keys. The recommended approach is to use .env files:

1. Create a .env file in your project root or in the location specified by your Goose configuration
2. Use the provided .env.example file as a reference:

   ANTHROPIC_API_KEY=your_anthropic_api_key_here
   GOOGLE_API_KEY=your_gemini_api_key_here
   

3. Copy .env.example to .env and add your actual API keys
4. Ensure your .env file is added to .gitignore to prevent accidentally committing sensitive information

Gemini CLI API Keys

For the Gemini CLI to work, you'll need to configure your Google AI API key:

1. Get your API key from Google AI Studio
2. Set it as an environment variable:

   GEMINI_API_KEY=your_gemini_api_key_here
   

3. Add this to your .env file or set it in your dev container environment

Available Options

• The goose feature doesn't currently take any options
• The gemini-cli feature doesn't currently take any options

Repository Structure

.
├── src
│   ├── goose                   # Goose Feature
│   │   ├── devcontainer-feature.json
│   │   └── install.sh
│   └── gemini                  # Gemini CLI Feature
│       ├── devcontainer-feature.json
│       └── install.sh
└── test                        # Tests for the Features
    ├── goose
    │   ├── test.json
    │   └── test.sh
    └── gemini
        ├── test.json
        └── test.sh

Development

Prerequisites

• Docker
• Dev Container CLI

Testing

To test the feature, use the Dev Container CLI:

devcontainer features test

This will build test containers with both features installed and run the tests defined in test/goose/test.sh and test/gemini/test.sh.

Publishing a New Release

To publish a new version of this feature:

1. Update the code as needed and test locally
2. Update the version in the appropriate feature's devcontainer-feature.json file (src/goose/devcontainer-feature.json or src/gemini/devcontainer-feature.json)
3. Commit your changes and push to the main branch
4. Create and push a new version tag:

   git tag v1.0.0  # Replace with your version
   git push origin v1.0.0

5. The GitHub Actions workflow will automatically:
   • Run all tests on multiple base images
   • Publish the feature to GitHub Container Registry if tests pass
   • Generate updated documentation and create a PR

After the release is published, users can install the feature using the reference in their devcontainer.json file.

Contributing

Contributions are welcome! Here's how you can contribute:

1. Fork the repository
2. Create a feature branch: git checkout -b my-new-feature
3. Make your changes
4. Test your changes: devcontainer features test (tests both features) or devcontainer features test -f goose / devcontainer features test -f gemini for individual features
5. Commit your changes: git commit -am 'Add some feature'
6. Push to the branch: git push origin my-new-feature
7. Submit a pull request

Building Locally for Testing in Other Projects

To build and test your feature in another project before publishing:

1. Create a tarball of your feature:

   devcontainer features package -f -o ./output ./src

   This creates a tarball in the ./output directory.

This approach allows you to test your feature changes in real projects before publishing a new version.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Block for creating the Goose AI coding assistant
• Google for creating the Gemini CLI
• Dev Containers for the dev container specification