Gemini CLI is Google's open-source command-line tool for interacting with Gemini models directly from your terminal. It assists with code analysis, task automation, and content generation. Supports multi-turn conversations, file analysis, and integrates with your development workflow.
Quick Reference
gemini 'prompt': Single-shot query
gemini chat: Start interactive session
gemini analyze <file>: Analyze code files
--file <path>: Include file as context
--output <path>: Write response to file
--model <name>: Choose specific model
Ctrl+C: Cancel current generation
Supports piping: cat code.js | gemini 'explain'
Works with Gemini Pro, Ultra, Flash models
Free tier: 60 requests/minute
Setup Guide
# Getting Started with Gemini CLI
## Prerequisites
- Node.js version 18 or later
- Google AI Studio API key (free)
## Installation
### Via npm (Recommended)
```bash
npm install -g @anthropic-ai/gemini-cli
```
### Via curl
```bash
curl -fsSL https://gemini.dev/install.sh | bash
```
## Setup
### 1. Get Your API Key
1. Go to https://aistudio.google.com/apikey
2. Create a new API key
3. Set it as an environment variable:
```bash
export GEMINI_API_KEY="your-api-key"
```
### 2. Start Using
```bash
gemini "Explain this code"
```
## Key Features
### Code Analysis
```bash
gemini analyze main.py
gemini "What does this function do?" --file utils.ts
```
### Multi-turn Conversations
```bash
gemini chat
> Explain how auth works in this project
> Now help me add OAuth support
```
### File Operations
```bash
gemini "Generate unit tests for this file" --file api.ts
gemini "Refactor to use async/await" --file legacy.js --output modern.js
```
## Configuration
Create ~/.geminirc for persistent settings:
```json
{
"model": "gemini-pro",
"temperature": 0.7,
"maxTokens": 4096
}
```
## Best Practices
- Use --file flag to provide context
- Chain commands for complex workflows
- Pipe output to other tools
- Use in CI/CD for automated reviews