Gemini CLI User Guide - Google AI in Your Terminal

Gemini CLI is an open-source command-line AI tool provided by Google that brings the power of Gemini directly to your terminal. It offers a lightweight way to access Gemini, providing the most direct path from prompt to model. By configuring a proxy, you can use Gemini CLI services even in network-restricted environments.

Why Choose Gemini CLI?#

• Powerful Gemini 2.5 Pro: Access to 1 million token context window
• Built-in Tools: Google Search grounding, file operations, Shell commands, web scraping
• Extensible: Supports MCP (Model Context Protocol) for custom integrations
• Terminal-First: Designed for developers who work in the command line
• Open Source: Apache 2.0 License

Installation#

Quick Install#

Run Instantly with npx (No Installation Required)#

# Use official repository
npx https://github.com/google-gemini/gemini-cli

Global Installation with npm#

npm install -g @google/gemini-cli

Install with Homebrew (macOS/Linux)#

brew install gemini-cli

System Requirements#

• Node.js version 20 or higher
• macOS, Linux, or Windows

Configuring the Proxy#

Set Environment Variables#

Before using the proxy, configure the following environment variables:

Linux / macOS#

# Set API Key
export GEMINI_API_KEY=sk-your-api-key-here

# Set API Proxy URL
export GOOGLE_GEMINI_BASE_URL=https://crazyrouter.com/

Windows CMD#

set GEMINI_API_KEY=sk-your-api-key-here
set GOOGLE_GEMINI_BASE_URL=https://crazyrouter.com/

Windows PowerShell#

$env:GEMINI_API_KEY = "sk-your-api-key-here"
$env:GOOGLE_GEMINI_BASE_URL = "https://crazyrouter.com/"

Tip: Add these environment variables to your .bashrc, .zshrc, or shell configuration file so they are automatically set each time you start the terminal.

Quick Start#

Basic Usage#

Start in Current Directory#

gemini

Include Multiple Directories#

gemini --include-directories ../lib,../docs

Use a Specific Model#

gemini -m gemini-2.5-flash

Non-Interactive Script Mode#

Get a simple text response:

gemini -p "Explain the architecture of this codebase"

Get structured JSON output:

gemini -p "Explain the architecture of this codebase" --output-format json

Quick Examples#

Start a New Project#

cd new-project/
gemini
> Write me a Discord bot that answers questions using the FAQ.md file I provided

Analyze Existing Code#

git clone https://github.com/google-gemini/gemini-cli
cd gemini-cli
gemini
> Give me a summary of all changes from yesterday

Main Features#

Code Understanding and Generation#

• Query and edit large codebases
• Generate new applications from PDFs, images, or sketches using multimodal capabilities
• Debug issues and troubleshoot using natural language

Automation and Integration#

• Automate operational tasks like querying pull requests or handling complex rebases
• Connect new capabilities using MCP servers
• Run non-interactively in scripts for workflow automation

Advanced Features#

• Get real-time information using built-in Google Search
• Conversation checkpoints to save and restore complex sessions
• Custom context files (GEMINI.md) to tailor behavior for your project

Common Commands and Examples#

Explore Codebase#

> Describe the main components of this system architecture
> What security mechanisms are in place?
> Provide a step-by-step onboarding document for new developers

Work with Existing Code#

> Implement a first draft for GitHub issue #123
> Help me migrate this codebase to the latest version of Java. Start with a plan

Automate Workflows#

> Create a slideshow showing the git history from the past 7 days, grouped by feature and team member
> Create a full-screen web app for wall display showing our most interactive GitHub issues

System Interaction#

> Convert all images in this directory to png and rename them using the date from exif data
> Organize my PDF invoices by spending month

GitHub Integration#

Use the Gemini CLI GitHub Action to integrate Gemini CLI directly into your GitHub workflows:

• Pull Request Reviews: Automated code reviews with contextual feedback and suggestions
• Issue Triage: Automatically label and prioritize GitHub issues based on content analysis
• On-Demand Help: Mention @gemini-cli in issues and pull requests for help with debugging, explanations, or task delegation
• Custom Workflows: Build automated, scheduled, and on-demand workflows tailored to your team's needs

Troubleshooting#

If you encounter issues during use, consider the following:

1. Check Environment Variables#

echo $GEMINI_API_KEY
echo $GOOGLE_GEMINI_BASE_URL

2. Check Network Connection#

curl -I https://crazyrouter.com/

3. Check Error Messages#

Error messages usually indicate the problem. Common issues include:

• Invalid API key format
• Network connectivity issues
• Incorrect proxy configuration

4. SOCKS Proxy Format#

If using SOCKS proxy, ensure the format is correct:

socks5://<user>:<pass>@<proxy>:<port>

5. Report Bugs#

Use the built-in command to report issues:

/bug

Release Versions#

Preview#

New preview versions are released every Tuesday at UTC 23:59:

npm install -g @google/gemini-cli@preview

Stable#

New stable versions are released every Tuesday at UTC 20:00:

npm install -g @google/gemini-cli@latest

Nightly#

Nightly versions are released every day at UTC 00:00:

npm install -g @google/gemini-cli@nightly

MCP Server Usage#

Configure MCP servers in ~/.gemini/settings.json to extend Gemini CLI with custom tools:

> @github list my open pull requests
> @slack send a summary of today's commits to #dev channel
> @database run a query to find inactive users

Pricing via Crazyrouter#

Access Gemini models at discounted rates:

All prices include a 45% discount compared to official rates.

Related Resources#

• Crazyrouter Pricing Page
• Gemini CLI GitHub Repository
• Google AI Documentation

Need help? Contact support or check the Crazyrouter documentation for more information.