MiCLI

Take XiaoMi Cloud Service to the command line.

MiCLI is a powerful CLI tool for controlling MIoT devices, XiaoAi speakers (Mina), and MiHome devices. It supports device property reading/writing, TTS synthesis, AI conversations, and more.

Features

• 🏠 MIoT Device Control - Read and write device properties
• 🔊 XiaoAi Speaker - TTS, playback control, voice records
• 🤖 AI Integration - ChatGPT integration for smart conversations
• 🎙️ Edge TTS - Microsoft Edge TTS without API key
• 📱 QR Code Login - Scan QR code with Mi Home app to authenticate
• 🌐 Web Server - Built-in Gin server for extended functionality
• ⚙️ Automation - Keyword-based command chaining via commands.json

Installation

Build from Source

git clone https://github.com/WangNingkai/micli.git
cd micli
go build -o micli

Prerequisites

• Go 1.25 or higher
• XiaoMi account

Quick Start

1. Initialize Configuration

Run any command for the first time, it will create conf.ini interactively:

./micli list

Configuration options:

• MI_USER / MI_PASS - XiaoMi account credentials
• REGION - Region (cn/de/us/i2/ru/sg/tw)
• MI_DID - Default device DID
• OpenAI settings (optional)

2. List Devices

./micli list

3. Device Properties

# Get device properties
./micli get -d <device_id> --props <prop1>,<prop2>

# Set device properties
./micli set -d <device_id> --props <prop1>=<value1>

Property format: siid-piid (e.g., 2-1 for service 2, property 1)

4. MIoT Actions

# Execute MIoT action
./micli action <device_id> <siid-aiid> [args...]

5. XiaoAi Speaker

# TTS synthesis
./micli mina tts -d <device_id> --text "Hello, World!"

# List XiaoAi devices
./micli mina list

# Player control (play/pause/volume)
./micli mina player <command>

# Get voice records
./micli mina records [limit]

# Set default XiaoAi device
./micli mina set_did

# Start AI conversation mode
./micli mina serve

6. Text-to-Speech

# Interactive voice selection
./micli tts -t "你好，世界"

# Specify voice
./micli tts -t "Hello" -v en-US-JennyNeural

7. QR Code Login

# Generate QR code for Mi Home app scan
./micli qr-login

Commands Reference

Command	Description
list	List all devices
get	Get MIoT device properties
set	Set MIoT device properties
action <iid> [args]	Execute MIoT action
spec [model]	Show MIoT specification
decode	Decode MIoT encrypted data
mina	XiaoAi speaker commands
mina list	List XiaoAi devices
mina tts	TTS for XiaoAi
mina player	Player control (play/pause/volume)
mina records [limit]	Get voice conversation records
mina serve	Start web server + AI conversation
mina set_did	Set default XiaoAi DID
tts -t <text>	Edge TTS synthesis
miot_raw <cmd> <params>	Raw MIoT API call
miio_raw <uri> <data>	Raw MiIO API call
set_did	Set default MIoT DID
qr-login	QR code authentication
reset	Reset configuration

Automation

Create commands.json (see commands.sample.json for examples) to define keyword-triggered automation chains:

[
  {
    "keyword": "天气怎么样",
    "step": [
      {
        "type": "request",
        "request": {
          "url": "http://www.weather.com.cn/data/sk/101010100.html",
          "out": "res.data",
          "wait": true
        }
      }
    ]
  },
  {
    "keyword": "开启高级对话",
    "type": "chat",
    "chat": {
      "endFlag": "关闭高级对话",
      "stream": true,
      "useEdgeTTS": false
    }
  }
]

Step types: tts, request, action, chat (ChatGPT)

MIoT Specification

Find your device specification at miot-spec.

Configuration

Configuration file: conf.ini

[app]
DEBUG = false
PORT = :8080

[account]
MI_USER = your_email@example.com
MI_PASS = your_password
REGION = cn
MI_DID =

[mina]
DID = your_device_did

[openai]
KEY = sk-xxx
BASE_URL = https://api.openai.com/v1

[file]
TRANSFER_SH = https://transfer.sh

Token cache: ~/.mi.token

Architecture

main.go → cmd.Execute() (Cobra)
           │
           ├─ cmd/               # Cobra commands
           │   ├─ root.go        # Config & service init
           │   ├─ mina*.go       # XiaoAi commands
           │   ├─ props_*.go     # MIoT operations
           │   ├─ action.go      # MIoT actions
           │   ├─ spec.go        # MIoT spec viewer
           │   ├─ miot_raw.go    # Raw MIoT API
           │   ├─ miio_raw.go    # Raw MiIO API
           │   ├─ tts.go         # Edge TTS command
           │   ├─ decode.go      # MIoT decoder
           │   ├─ set_did.go     # Set default DID
           │   ├─ qr_login.go    # QR authentication
           │   └─ reset.go       # Reset config
           │
           ├─ pkg/
           │   ├─ miservice/     # XiaoMi API core
           │   │   ├─ service.go # Auth & requests
           │   │   ├─ mina.go    # XiaoAi API
           │   │   ├─ io.go      # MIoT/MiIO ops
           │   │   ├─ token.go   # Token storage
           │   │   └─ qrlogin.go # QR login
           │   ├─ jarvis/        # ChatGPT integration
           │   ├─ tts/           # Edge TTS
           │   └─ util/          # Utilities (signing, crypto)
           │
           ├─ internal/
           │   ├─ app.go         # Gin server
           │   ├─ conf/          # Configuration
           │   ├─ handlers/      # HTTP handlers
           │   ├─ middleware/    # CORS middleware
           │   └─ static/        # Static files
           │
           ├─ public/            # Embedded frontend (embed.FS)
           │   └─ dist/          # Frontend build output
           │
           └─ data/              # Local cache
               ├─ devices.json   # Device list cache
               └─ miot-spec.json # MIoT spec cache

Development

# Run
go run main.go [command]

# Build
go build -o micli

# Test
go test ./...

# Test with race detection
go test -race ./...

# Lint
golangci-lint run

CI/CD

GitHub Actions builds binaries for Linux (amd64/arm64), macOS, and Windows on version tags (excluding -alpha*).

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Cobra - CLI framework
• Gin - Web framework
• go-openai - OpenAI API client
• pterm - Terminal UI
• req - HTTP client
• Edge TTS - Microsoft Edge TTS

---

Disclaimer: This project is for educational purposes only. Use at your own risk.