@iflow-ai/iflow-cli

🤖 iFlow CLI

English | 中文 | 日本語 | 한국어 | Français | Deutsch | Español | Русский

iFlow CLI is a powerful AI assistant that runs directly in your terminal. It seamlessly analyzes code repositories, executes coding tasks, understands context-specific needs, and boosts productivity by automating everything from simple file operations to complex workflows.

✨ Key Features

• Free AI Models: Access powerful and free AI models through the iFlow open platform, including Kimi K2, Qwen3 Coder, DeepSeek v3, and more
• Flexible Integration: Full support for OpenAI protocol model providers
• Intuitive Interface: Streamlined terminal experience with context-aware assistance
• Ready-to-Use Assistant: Pre-configured MCP servers and specialized agents that work together to solve complex problems right out of the box

📥 Installation

System requirements

• Operating Systems: macOS 10.15+, Ubuntu 20.04+/Debian 10+, or Windows 10+ (with WSL 1, WSL 2, or Git for Windows)
• Hardware: 4GB+ RAM
• Software: Node.js 18+
• Network: Internet connection required for authentication and AI processing
• Shell: Works best in Bash, Zsh or Fish

Install Command

bash -c "$(curl -fsSL https://cloud.iflow.cn/iflow-cli/install.sh)"

This command automatically installs all necessary dependencies for your terminal.

Windows Users:

• Go to https://nodejs.org/en/download to download the latest Node.js installer
• Run the installer to install Node.js
• Restart your terminal: CMD or PowerShell
• Run npm install -g @iflow-ai/iflow-cli to install iFlow CLI
• Run iflow to start iFlow CLI

If you are in China Mainland, you can use the following command to install iFlow CLI:

• Go to https://cloud.iflow.cn/iflow-cli/nvm-setup.exe to download the latest nvm installer
• Run the installer to install nvm
• Restart your terminal: CMD or PowerShell
• Run nvm node_mirror https://npmmirror.com/mirrors/node/ and nvm npm_mirror https://npmmirror.com/mirrors/npm/
• Run nvm install 22 to install Node.js 22
• Run nvm use 22 to use Node.js 22
• Run npm install -g @iflow-ai/iflow-cli to install iFlow CLI
• Run iflow to start iFlow CLI

🔑 Authentication

iFlow offers two authentication options:

• Recommended: Use iFlow's native authentication
• Alternative: Connect via OpenAI-compatible APIs

To get your API key:

• Register for an iFlow account
• Go to your profile settings or click this direct link
• Click "Reset" in the pop-up dialog to generate a new API key

After generating your key, paste it into the terminal prompt to complete setup.

🚀 Getting Started

To launch iFlow CLI, navigate to your workspace in the terminal and type:

iflow

Starting New Projects

For new projects, simply describe what you want to create:

cd new-project/
iflow
> Create a web-based Minecraft game using HTML

Working with Existing Projects

For existing codebases, begin with the /init command to help iFlow understand your project:

cd project1/
iflow
> /init
> Analyze the requirements according to the PRD document in requirement.md file, and output a technical document, then implement the solution.

The /init command scans your codebase, learns its structure, and creates an IFLOW.md file with comprehensive documentation.

For a complete list of slash commands and usage instructions, see here.

💡 Common Use Cases

iFlow CLI extends beyond coding to handle a wide range of tasks:

📊 Information & Planning

> Help me find the best-rated restaurants in Los Angeles and create a 3-day food tour itinerary.

> Search for the latest iPhone price comparisons and find the most cost-effective purchase option.

📁 File Management

> Organize the files on my desktop by file type into separate folders.

> Batch download all images from this webpage and rename them by date.

📈 Data Analysis

> Analyze the sales data in this Excel spreadsheet and generate a simple chart.

> Extract customer information from these CSV files and merge them into a unified table.

👨‍💻 Development Support

> Analyze the main architectural components and module dependencies of this system.

> I'm getting a null pointer exception after my request, please help me find the cause of the problem.

⚙️ Workflow Automation

> Create a script to periodically backup my important files to cloud storage.

> Write a program that downloads stock prices daily and sends me email notifications.

Note: Advanced automation tasks can leverage MCP servers to integrate your local system tools with enterprise collaboration suites.

🔧 Switch to Custom Model

iFlow CLI can connect to any OpenAI-compatible API. Edit the settings file in ~/.iflow/settings.json to change the model you use.

Here is a settings demo file:

{
    "theme": "Default",
    "selectedAuthType": "iflow",
    "apiKey": "your iflow key",
    "baseUrl": "https://apis.iflow.cn/v1",
    "modelName": "Qwen3-Coder",
    "searchApiKey": "your iflow key"
}

🐛 Crash Debugging

If you encounter runtime crashes, you can enable crash diagnostics to collect detailed information for debugging:

Enable Crash Diagnostics

# Enable crash debugging mode
DEBUG=true iflow

# Or with additional Node.js debugging flags
DEBUG=true node --trace-warnings --trace-uncaught node_modules/.bin/iflow

System Configuration for Core Dumps

macOS:

# Enable core dumps
ulimit -c unlimited
sudo sysctl -w kern.coredump=1

Linux:

# Configure core dump pattern and enable unlimited core dumps
echo 'core.%p.%t' | sudo tee /proc/sys/kernel/core_pattern
ulimit -c unlimited

Crash Information Collection

When DEBUG=true is enabled, the system will:

• Generate crash files - Create crash-[timestamp]-[pid].json files containing:

  • Complete error stack trace
  • Memory usage information
  • Process information (PID, uptime, Node.js version)
  • Environment variables
  • Platform and architecture details

• Create core dumps - Generate system core dumps for advanced debugging

• Monitor warnings - Log Node.js warnings and unhandled promise rejections

Analyzing Crash Files

The generated crash files contain structured JSON data that can help identify:

• The exact error that caused the crash
• Memory usage patterns before the crash
• System environment and configuration
• Call stack at the time of crash

Share these files when reporting crash issues for faster debugging assistance.

GitHub Actions

You can also use iFlow CLI in your GitHub Actions workflows with the community-maintained action: iflow-cli-action

Community Communication

If you encounter problems during use, you can directly raise Issues on the GitHub page.

You can also scan the following WeChat group to join the community group for communication and discussion.

WeChat Group

1235 1235