#!/usr/bin/env python3
"""
Phone Agent CLI - AI-powered phone automation.
Usage:
python main.py [OPTIONS]
Environment Variables:
PHONE_AGENT_BASE_URL: Model API base URL (default: http:
PHONE_AGENT_MODEL: Model name (default: autoglm-phone-9b)
PHONE_AGENT_API_KEY: API key for model authentication (default: EMPTY)
PHONE_AGENT_MAX_STEPS: Maximum steps per task (default: 100)
PHONE_AGENT_DEVICE_ID: ADB device ID for multi-device setups
"""
import os
import shutil
import subprocess
import sys
from urllib.parse import urlparse
from openai import OpenAI
from phone_agent import PhoneAgent
from phone_agent.agent import AgentConfig
from phone_agent.agent_ios import IOSAgentConfig, IOSPhoneAgent
from phone_agent.config.apps import list_supported_apps
from phone_agent.config.apps_harmonyos import list_supported_apps as list_harmonyos_apps
from phone_agent.config.apps_ios import list_supported_apps as list_ios_apps
from phone_agent.device_factory import DeviceType, get_device_factory, set_device_type
from phone_agent.model import ModelConfig
from phone_agent.xctest import XCTestConnection
from phone_agent.xctest import list_devices as list_ios_devices
def check_system_requirements(
device_type: DeviceType = DeviceType.ADB, wda_url: str = "http://localhost:8100"
) -> bool:
"""
Check system requirements before running the agent.
Checks:
1. ADB/HDC/iOS tools installed
2. At least one device connected
3. ADB Keyboard installed on the device (for ADB only)
4. WebDriverAgent running (for iOS only)
Args:
device_type: Type of device tool (ADB, HDC, or IOS).
wda_url: WebDriverAgent URL (for iOS only).
Returns:
True if all checks pass, False otherwise.
"""
print("🔍 Checking system requirements...")
print("-" * 50)
all_passed = True
# Determine tool name and command
if device_type == DeviceType.IOS:
tool_name = "libimobiledevice"
tool_cmd = "idevice_id"
else:
tool_name = "ADB" if device_type == DeviceType.ADB else "HDC"
tool_cmd = "adb" if device_type == DeviceType.ADB else "hdc"
# Check 1: Tool installed
print(f"1. Checking {tool_name} installation...", end=" ")
if shutil.which(tool_cmd) is None:
print("❌ FAILED")
print(f" Error: {tool_name} is not installed or not in PATH.")
print(f" Solution: Install {tool_name}:")
if device_type == DeviceType.ADB:
print(" - macOS: brew install android-platform-tools")
print(" - Linux: sudo apt install android-tools-adb")
print(
" - Windows: Download from https://developer.android.com/studio/releases/platform-tools"
)
elif device_type == DeviceType.HDC:
print(
" - Download from HarmonyOS SDK or https://gitee.com/openharmony/docs"
)
print(" - Add to PATH environment variable")
else: # IOS
print(" - macOS: brew install libimobiledevice")
print(" - Linux: sudo apt-get install libimobiledevice-utils")
all_passed = False
else:
# Double check by running version command
try:
if device_type == DeviceType.ADB:
version_cmd = [tool_cmd, "version"]
elif device_type == DeviceType.HDC:
version_cmd = [tool_cmd, "-v"]
else: # IOS
version_cmd = [tool_cmd, "-ln"]
result = subprocess.run(
version_cmd, capture_output=True, text=True, timeout=10
)
if result.returncode == 0:
version_line = result.stdout.strip().split("\n")[0]
print(f"✅ OK ({version_line if version_line else 'installed'})")
else:
print("❌ FAILED")
print(f" Error: {tool_name} command failed to run.")
all_passed = False
except FileNotFoundError:
print("❌ FAILED")
print(f" Error: {tool_name} command not found.")
all_passed = False
except subprocess.TimeoutExpired:
print("❌ FAILED")
print(f" Error: {tool_name} command timed out.")
all_passed = False
# If ADB is not installed, skip remaining checks
if not all_passed:
print("-" * 50)
print("❌ System check failed. Please fix the issues above.")
return False
# Check 2: Device connected
print("2. Checking connected devices...", end=" ")
try:
if device_type == DeviceType.ADB:
result = subprocess.run(
["adb", "devices"], capture_output=True, text=True, timeout=10
)
lines = result.stdout.strip().split("\n")
# Filter out header and empty lines, look for 'device' status
devices = [
line for line in lines[1:] if line.strip() and "\tdevice" in line
]
elif device_type == DeviceType.HDC:
result = subprocess.run(
["hdc", "list", "targets"], capture_output=True, text=True, timeout=10
)
lines = result.stdout.strip().split("\n")
devices = [line for line in lines if line.strip()]
else: # IOS
ios_devices = list_ios_devices()
devices = [d.device_id for d in ios_devices]
if not devices:
print("❌ FAILED")
print(" Error: No devices connected.")
print(" Solution:")
if device_type == DeviceType.ADB:
print(" 1. Enable USB debugging on your Android device")
print(" 2. Connect via USB and authorize the connection")
print(
" 3. Or connect remotely: python main.py --connect <ip>:<port>"
)
elif device_type == DeviceType.HDC:
print(" 1. Enable USB debugging on your HarmonyOS device")
print(" 2. Connect via USB and authorize the connection")
print(
" 3. Or connect remotely: python main.py --device-type hdc --connect <ip>:<port>"
)
else: # IOS
print(" 1. Connect your iOS device via USB")
print(" 2. Unlock device and tap 'Trust This Computer'")
print(" 3. Verify: idevice_id -l")
print(" 4. Or connect via WiFi using device IP")
all_passed = False
else:
if device_type == DeviceType.ADB:
device_ids = [d.split("\t")[0] for d in devices]
elif device_type == DeviceType.HDC:
device_ids = [d.strip() for d in devices]
else: # IOS
device_ids = devices
print(
f"✅ OK ({len(devices)} device(s): {', '.join(device_ids[:2])}{'...' if len(device_ids) > 2 else ''})"
)
except subprocess.TimeoutExpired:
print("❌ FAILED")
print(f" Error: {tool_name} command timed out.")
all_passed = False
except Exception as e:
print("❌ FAILED")
print(f" Error: {e}")
all_passed = False
# If no device connected, skip ADB Keyboard check
if not all_passed:
print("-" * 50)
print("❌ System check failed. Please fix the issues above.")
return False
# Check 3: ADB Keyboard installed (only for ADB) or WebDriverAgent (for iOS)
if device_type == DeviceType.ADB:
print("3. Checking ADB Keyboard...", end=" ")
try:
result = subprocess.run(
["adb", "shell", "ime", "list", "-s"],
capture_output=True,
text=True,
timeout=10,
)
ime_list = result.stdout.strip()
if "com.android.adbkeyboard/.AdbIME" in ime_list:
print("✅ OK")
else:
print("❌ FAILED")
print(" Error: ADB Keyboard is not installed on the device.")
print(" Solution:")
print(" 1. Download ADB Keyboard APK from:")
print(
" https://github.com/senzhk/ADBKeyBoard/blob/master/ADBKeyboard.apk"
)
print(" 2. Install it on your device: adb install ADBKeyboard.apk")
print(
" 3. Enable it in Settings > System > Languages & Input > Virtual Keyboard"
)
all_passed = False
except subprocess.TimeoutExpired:
print("❌ FAILED")
print(" Error: ADB command timed out.")
all_passed = False
except Exception as e:
print("❌ FAILED")
print(f" Error: {e}")
all_passed = False
elif device_type == DeviceType.HDC:
# For HDC, skip keyboard check as it uses different input method
print("3. Skipping keyboard check for HarmonyOS...", end=" ")
print("✅ OK (using native input)")
else: # IOS
# Check WebDriverAgent
print(f"3. Checking WebDriverAgent ({wda_url})...", end=" ")
try:
conn = XCTestConnection(wda_url=wda_url)
if conn.is_wda_ready():
print("✅ OK")
# Get WDA status for additional info
status = conn.get_wda_status()
if status:
session_id = status.get("sessionId", "N/A")
print(f" Session ID: {session_id}")
else:
print("❌ FAILED")
print(" Error: WebDriverAgent is not running or not accessible.")
print(" Solution:")
print(" 1. Run WebDriverAgent on your iOS device via Xcode")
print(" 2. For USB: Set up port forwarding: iproxy 8100 8100")
print(
" 3. For WiFi: Use device IP, e.g., --wda-url http://192.168.1.100:8100"
)
print(" 4. Verify in browser: open http://localhost:8100/status")
all_passed = False
except Exception as e:
print("❌ FAILED")
print(f" Error: {e}")
all_passed = False
print("-" * 50)
if all_passed:
print("✅ All system checks passed!\n")
else:
print("❌ System check failed. Please fix the issues above.")
return all_passed
def check_model_api(base_url: str, model_name: str, api_key: str = "EMPTY") -> bool:
"""
Check if the model API is accessible and the specified model exists.
Checks:
1. Network connectivity to the API endpoint
2. Model exists in the available models list
Args:
base_url: The API base URL
model_name: The model name to check
api_key: The API key for authentication
Returns:
True if all checks pass, False otherwise.
"""
print("🔍 Checking model API...")
print("-" * 50)
all_passed = True
# Check 1: Network connectivity using chat API
print(f"1. Checking API connectivity ({base_url})...", end=" ")
try:
# Create OpenAI client
client = OpenAI(base_url=base_url, api_key=api_key, timeout=30.0)
# Use chat completion to test connectivity (more universally supported than /models)
response = client.chat.completions.create(
model=model_name,
messages=[{"role": "user", "content": "Hi"}],
max_tokens=5,
temperature=0.0,
stream=False,
)
# Check if we got a valid response
if response.choices and len(response.choices) > 0:
print("✅ OK")
else:
print("❌ FAILED")
print(" Error: Received empty response from API")
all_passed = False
except Exception as e:
print("❌ FAILED")
error_msg = str(e)
# Provide more specific error messages
if "Connection refused" in error_msg or "Connection error" in error_msg:
print(f" Error: Cannot connect to {base_url}")
print(" Solution:")
print(" 1. Check if the model server is running")
print(" 2. Verify the base URL is correct")
print(f" 3. Try: curl {base_url}/chat/completions")
elif "timed out" in error_msg.lower() or "timeout" in error_msg.lower():
print(f" Error: Connection to {base_url} timed out")
print(" Solution:")
print(" 1. Check your network connection")
print(" 2. Verify the server is responding")
elif (
"Name or service not known" in error_msg
or "nodename nor servname" in error_msg
):
print(f" Error: Cannot resolve hostname")
print(" Solution:")
print(" 1. Check the URL is correct")
print(" 2. Verify DNS settings")
else:
print(f" Error: {error_msg}")
all_passed = False
print("-" * 50)
if all_passed:
print("✅ Model API checks passed!\n")
else:
print("❌ Model API check failed. Please fix the issues above.")
return all_passed
def main():
"""Main entry point."""
import json
# Read configuration from file
with open('config.json', 'r', encoding='utf-8') as f:
config = json.load(f)
base_url = config.get('base_url', 'https://open.bigmodel.cn/api/paas/v4')
model = config.get('model', 'autoglm-phone')
apikey = config.get('apikey', 'abcd7090c77')
connect_addr = config.get('connect', '127.0.0.1:5555') # Address to connect to remote device
# Handle task which can be either a string or an array of strings
task_config = config.get('task', '打开百度')
if isinstance(task_config, list):
task = ', '.join(task_config) # Join array elements with comma
else:
task = task_config # Use as is if it's already a string
# Define fixed parameters
device_type_str = "adb" # Default to ADB, change to "ios" or "hdc" as needed
device_id = None # Use default device detection
max_steps = 100
lang = "cn" # Default language
quiet = False # Verbose output
wda_url = "http://localhost:8100" # Default WDA URL for iOS
# Set device type globally based on fixed value
if device_type_str == "adb":
device_type = DeviceType.ADB
elif device_type_str == "hdc":
device_type = DeviceType.HDC
else: # ios
device_type = DeviceType.IOS
# Set device type globally for non-iOS devices
if device_type != DeviceType.IOS:
set_device_type(device_type)
# Enable HDC verbose mode if using HDC
if device_type == DeviceType.HDC:
from phone_agent.hdc import set_hdc_verbose
set_hdc_verbose(True)
# Handle --list-apps (no system check needed)
# Skip this functionality since it's command-line specific
# Handle device commands (these may need partial system checks)
# Skip these since they're command-line specific
# Connect to remote device if connect address is provided
if connect_addr:
print(f"Connecting to remote device at {connect_addr}...")
device_factory = get_device_factory()
ConnectionClass = device_factory.get_connection_class()
conn = ConnectionClass()
success, message = conn.connect(connect_addr)
print(f"{'✓' if success else '✗'} {message}")
if success:
# Set as default device
device_id = connect_addr
# Run system requirements check before proceeding
if not check_system_requirements(
device_type,
wda_url=wda_url
if device_type == DeviceType.IOS
else "http://localhost:8100",
):
sys.exit(1)
# Check model API connectivity and model availability
if not check_model_api(base_url, model, apikey):
sys.exit(1)
# Create configurations and agent based on device type
model_config = ModelConfig(
base_url=base_url,
model_name=model,
api_key=apikey,
lang=lang,
)
if device_type == DeviceType.IOS:
# Create iOS agent
agent_config = IOSAgentConfig(
max_steps=max_steps,
wda_url=wda_url,
device_id=device_id,
verbose=not quiet,
lang=lang,
)
agent = IOSPhoneAgent(
model_config=model_config,
agent_config=agent_config,
)
else:
# Create Android/HarmonyOS agent
agent_config = AgentConfig(
max_steps=max_steps,
device_id=device_id,
verbose=not quiet,
lang=lang,
)
agent = PhoneAgent(
model_config=model_config,
agent_config=agent_config,
)
# Print header
print("=" * 50)
if device_type == DeviceType.IOS:
print("Phone Agent iOS - AI-powered iOS automation")
else:
print("Phone Agent - AI-powered phone automation")
print("=" * 50)
print(f"Model: {model_config.model_name}")
print(f"Base URL: {model_config.base_url}")
print(f"Max Steps: {agent_config.max_steps}")
print(f"Language: {agent_config.lang}")
print(f"Device Type: {device_type_str.upper()}")
# Show iOS-specific config
if device_type == DeviceType.IOS:
print(f"WDA URL: {wda_url}")
# Show device info
if device_type == DeviceType.IOS:
devices = list_ios_devices()
if device_id:
print(f"Device: {device_id}")
elif devices:
device = devices[0]
print(f"Device: {device.device_name or device.device_id[:16]}")
if device.model and device.ios_version:
print(f" {device.model}, iOS {device.ios_version}")
else:
device_factory = get_device_factory()
devices = device_factory.list_devices()
if device_id:
print(f"Device: {device_id}")
elif devices:
print(f"Device: {devices[0].device_id} (auto-detected)")
print("=" * 50)
# Run with predefined task
task_config = config.get('task', '打开百度')
if isinstance(task_config, list):
print("\nTask:")
for i, step in enumerate(task_config, 1):
print(f" {i}. {step}")
task_str = ', '.join(task_config)
else:
print(f"\nTask: {task_config}\n")
task_str = task_config
result = agent.run(task_str)
print(f"\nResult: {result}")
if __name__ == "__main__":
main()
Config.json
{
"connect": "192.168.1.222:5555",
"base_url": "https://open.bigmodel.cn/api/paas/v4",
"model": "autoglm-phone",
"apikey": "5343c77e2264c0",
"task": [
"打开百度", "等待1秒",
"然后搜索ai agent", "等待1秒",
]
}
