Android — Direct over adb

Qirabot's built-in Android backend needs no Appium server, no framework, and nothing installed on the device for input: it shells out to adb (screenshot via screencap, input via input tap/swipe/keyevent). If adb devices sees it, Qirabot can drive it — real device or emulator.

Because element location is AI vision on the screenshot, there are no UiAutomator selectors and no accessibility-tree dependency: native apps, WebViews, Flutter, React Native, and games all look the same to it.

The core package is enough — no extras (see Installation). The quickest check is the CLI:

qirabot android "Open settings and turn on airplane mode"
qirabot android "..." -d emulator-5554 --app-package com.android.settings

The same thing in Python:

from qirabot import AdbDevice, Qirabot

device = AdbDevice()                 # or AdbDevice(serial="emulator-5554")
bot = Qirabot().bind(device)

bot.click("Login button")            # AI-located — no template images
result = bot.ai("Open Settings and turn on dark mode")
print(f"Success: {result.success}")
bot.close()

bind(device) fixes the target once, so every later call drops the first argument (bot.click("...") instead of bot.click(device, "...")) — see Custom Adapters & Bolt-On for the details.

Non-ASCII typing (Chinese, emoji)

Typing beyond ASCII works through the bundled ADBKeyboard IME — installed on demand and switched back afterwards. bot.type_text(...) with Chinese or emoji just works.

Screen recording

Record the device screen (not the host) into the run report:

bot = Qirabot(record_device=True)   # or QIRA_RECORD_DEVICE=1
bot.ai(device, "open settings")
bot.close()                         # pulls the video into report_dir/recording.mp4

Under the hood it's adb screenrecord; runs longer than screenrecord's 3-minute cap are merged with ffmpeg. From the CLI: qirabot android "..." --record.

Through Appium instead

Have an existing Appium setup or a cloud device farm? The same API drives an Appium driver — install qirabot[appium] and pass the driver:

from appium import webdriver
from appium.options.android import UiAutomator2Options
from qirabot import Qirabot

options = UiAutomator2Options()
options.platform_name = "Android"
options.device_name = "emulator-5554"
driver = webdriver.Remote("http://localhost:4723", options=options)
bot = Qirabot().bind(driver)

result = bot.ai("Open Display settings and change font size to Large")
bot.close()
driver.quit()

On the CLI, passing --appium-url selects the Appium engine: qirabot android "..." --appium-url http://localhost:4723. The full Appium workflow — device clouds, recording, and an Appium-vs-built-in comparison — is in Appium + Qirabot.

Platform notes

• press_key("Back") / "Home" / "Menu" map to adb keyevents; go_back sends keyevent BACK.
• long_press is available (touch platforms); hover is a no-op, right_click degrades to a tap.
• clear_text over raw adb is best-effort (caret-to-end + repeated delete) — there is no element model on purpose.
• Coming from Airtest 1.x? connect_device("Android:///emu-5554") becomes AdbDevice("emu-5554") — the rest of your bind() code is unchanged.
• Full per-action behavior: platform support matrix.