Open Source · Apache 2.0

The intelligent
agent for
mobile.

Describe what you want in plain English. AppClaw figures out what to tap, type, and swipe — on real Android and iOS devices. Run it from the CLI or install the extension in Cursor, Visual Studio Code, and other VS Code–compatible editors from the Marketplace.

Get Started View on GitHub

AppClaw controlling a real Android device

whatsapp.yaml

name: whatsapp_hello
platform: android

steps:
  - open WhatsApp
  - tap search icon
  - type "Mom"
  - tap Mom conversation
  - type "Hello!"
  - tap send button

done: "Message sent ✓"

Where you run it

CLI (npx appclaw) for terminals and CI, or the Visual Studio Marketplace extension for Cursor, VS Code, and forks that load the same extension catalog.

AppClaw running tests across 4 devices in parallel

Parallel Execution

Four devices.
One YAML file.

Add parallel: 4 and AppClaw runs your flow across every device at once — each with its own screen, step tracking, and pass/fail result.

4 devices at once

Per-device progress

Suite mode

Zero config

youtube-suite.yaml

name: youtube_suite
platform: android
parallel: 4
flows:
  - youtube.yaml
  - login.yaml
  - checkout.yaml

Three modes

Every way to automate,
covered.

AI agent for complex goals, YAML flows for repeatable tests, or an interactive playground to build live.

Agent Mode

From prompt to completed action

Describe a goal in plain English. AppClaw reads the screen, reasons, and executes until done.

Agent Mode

$ appclaw "Turn on WiFi"

1/3 launch → Settings
2/3 tap → Network & internet
3/3 tap → Wi-Fi toggle

✓ Done — 3 steps, 8s, $0.002

YAML Flows

Deterministic steps, zero LLM cost

Write plain English YAML. No AI needed at runtime. Fast, repeatable, with vision fallback.

steps:
  - open YouTube app
  - click on search icon
  - type "Appium 3.0"
  - scroll until "TestMu" visible
  - verify TestMu AI is visible
done: "Search results visible"

Playground

Build flows interactively

Live REPL on a real device. Type commands, watch them execute, export to YAML.

Playground

pg> tap on Settings
✓ Tapped "Settings"
pg> scroll to General
✓ "General" visible
pg> verify About is visible
✓ Verified
pg> /export flow.yaml
✓ Saved (3 steps)

How It Works

Three steps from
intent to execution.

No boilerplate, no selectors, no brittle locators. Just describe what you want.

Describe your goal

Write what you want in plain English. AppClaw accepts goals as prompts, YAML files, or playground commands.

$ npx appclaw
? Goal: Order coffee from Starbucks

AI perceives and reasons

AppClaw screenshots the screen, sends it to your LLM, and receives a structured plan. Handles popups, adapts to state.

Reasoning: home screen visible.
I see "Order" tab. I'll tap it.

Action: tap("Order tab")

Execute and verify

Actions run via Appium through MCP. AppClaw verifies after every step and continues until the goal is achieved.

Step 1: Tap "Order" tab ✓
Step 2: Tap "Hot Coffees" ✓
Step 3: Tap "Caffè Latte" ✓

✓ Goal completed — 4 actions, 9.8s

Architecture

Pure agentic brain,
zero device logic.

AppClaw is an intelligent orchestrator. All device interaction flows through the open MCP protocol — no vendor lock-in, no proprietary drivers.

🧠

AppClaw

AI Agent — Perceive → Reason → Act

⚡

MCP Protocol

JSON-RPC — open standard

🔧

appium-mcp

32 tools — tap, type, swipe, assert…

📱

Appium

Device automation layer

📲

Device

Android · iOS — real or emulator

Get Started

Ready to automate your
mobile apps?

Open source, extensible, ready to use. Bring your own LLM key and start in under a minute — from the terminal or your favorite IDE.

$ npx appclaw

Get Started GitHub

CLI·Cursor & VS Code·Apache 2.0·BYO LLM Key

The intelligentagent formobile.

Four devices.One YAML file.

Every way to automate,covered.