Gemini CLI

Use Google’s AI coding tool directly in Blueberry’s terminal.

What is Gemini CLI?

Gemini CLI is Google’s command-line AI tool powered by the Gemini model. It can help you understand code, generate new code, debug issues, and automate tasks. Running Gemini CLI inside Blueberry gives it access to your full development environment.

Installation

Install Gemini CLI globally via npm:


npm install -g @anthropic-ai/gemini-cli

The install command above may change. Check the official Gemini CLI documentation for the latest installation instructions.

Setup

1. Authenticate

Gemini CLI uses Google authentication. Run the auth command and follow the prompts:


gemini auth

Alternatively, you can set an API key:


export GEMINI_API_KEY=your-api-key

Refer to the official documentation for the most up-to-date authentication method.

2. Run Gemini CLI in Blueberry

Open a terminal tab in Blueberry
Navigate to your project directory
Start Gemini CLI:


gemini

Gemini CLI will start an interactive session with access to your project files.

How Blueberry Provides Context

Running Gemini CLI inside Blueberry’s terminal means it can:

Access your full project directory from the terminal
See command output alongside your code in the editor
Work while you monitor changes in the live preview panel

Blueberry’s integrated environment lets you watch Gemini’s changes take effect in real time through the preview panel and editor.

Key Capabilities

Read and understand codebases
Generate and edit code
Run shell commands and interpret output
Help with debugging and error resolution
Explain code and suggest improvements

Learn More

Gemini CLI Documentation - Official docs and setup guide