The skill is an instruction file for the agent plus a zero-dependency CLI. Once installed, your agent knows how to connect, tap and read the screen on its own.
Nothing to pick and nothing to unzip — send the line below to the agent you already use and it installs the skill for you.
In Claude Code, Codex, WorkBuddy or any other agent, just type:
Install the skill: https://xphoneuse.com/static/skills/xphoneuse.zip
Verify: ask the agent to run python scripts/xphoneuse-cli.py list — if it lists your device, you are done.
Prerequisite: the XPhoneUse desktop app is running with a phone connected.
Bundle with SKILL.md and the CLI · 24 KB · 2026-10-15
It covers the common actions on both Android and iPhone; Android additionally gets the UI tree and app management.
| Capability | Android | iPhone |
|---|---|---|
| List devices, launch apps | ✓ | ✓ |
| Tap and long-press | ✓ | ✓ |
| Swipe, by element or coordinates | ✓ | ✓ |
| Scroll lists and pages | ✓ | ✓ |
| Type text, including non-ASCII | ✓ | ✓ |
| Keys: back, home, enter | ✓ | ◐ |
| Screenshots with pixel diff | ✓ | ✓ |
| Read on-screen text with offline OCR | ✓ | ✓ |
| Screen recording, macro replay | ✓ | ✓ |
| Record and replay a flow | ✓ | ✓ |
| Read the UI tree | ✓ | — |
| Find an element by its text | ✓ | — |
| Wi-Fi debugging | ✓ | — |
| Rotate the screen | ✓ | — |
| Audio forwarding | ✓ | — |
| Browse and transfer files | ✓ | ◐ |
| Simulated location from a map | ✓ | ✓ |