This GitHub Action integrates the Gemini CLI into development workflows, enabling autonomous agents for routine coding tasks and on-demand collaboration for code reviews, issue triage, and analysis. It targets developers and teams seeking to leverage AI for enhanced productivity and code quality within their GitHub repositories.

How It Works

The action leverages the Gemini CLI, allowing conversational interaction via comments (e.g., @gemini-cli fix this issue). It supports tool-calling capabilities, enabling integration with other CLIs like gh for extended functionality. Customization is achieved through a GEMINI.md file, providing project-specific context and instructions to the Gemini model.

Quick Start & Requirements

1. Gemini API Key: Obtain from Google AI Studio and add as a GitHub Secret named GEMINI_API_KEY.
2. Workflow Setup: Use the /setup-github command within the Gemini CLI or manually copy example workflows to .github/workflows/.
3. Authentication: Supports Gemini API Key or Google Cloud Workload Identity Federation for Google services, and GITHUB_TOKEN or a custom GitHub App for GitHub API.
4. Dependencies: Requires a Gemini API key. Optional Google Cloud project details for Vertex AI or enhanced observability.

Highlighted Details

• Automated and on-demand triggering of workflows based on GitHub events or comments.
• Extensible with tool-calling for integration with other CLIs.
• Customizable with GEMINI.md for project-specific instructions.
• Supports observability by sending telemetry to a Google Cloud project.

Maintenance & Community

This project is maintained by Google. Further details on contributing can be found in the Gemini CLI Contributing Guide.

Licensing & Compatibility

The README does not explicitly state the license. Compatibility for commercial use or closed-source linking is not specified.

Limitations & Caveats

The README does not specify the license, which may impact commercial use. Authentication methods and observability configurations require careful setup.