burner-phone

Control Android devices via ADB commands with AI vision for screen analysis.

601|121|Updated Feb 5, 2026
One-click install
npx skills add https://github.com/smallnest/goclaw --skill burner-phone
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: burner-phone
Source: https://github.com/smallnest/goclaw/tree/main/internal/builtin_skills/burner-phone
Command: npx skills add https://github.com/smallnest/goclaw --skill burner-phone

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires adb, ollama, and includes scripts (resource) and assets (resource) components.

What problem does it solve?

This skill allows for direct control of Android devices using ADB commands, enhanced by AI vision to analyze screen content and perform precise actions.

Core Features & Use Cases

  • Vision-First Control: Leverages AI to interpret screen content and identify exact coordinates for interaction.
  • Direct Device Manipulation: Executes ADB commands for tapping, swiping, and typing directly on the device.
  • Use Case: Automate repetitive tasks on an Android device, such as navigating app interfaces, filling out forms within apps, or performing automated testing.

Quick Start

Use the burner-phone skill to take a screenshot of the current Android device screen.

Frequently Asked Questions about burner-phone

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate Android screen tapping and swiping using AI vision?

You can automate Android screen tapping and swiping by using ADB commands integrated with an AI vision model to interpret screen content and identify exact interaction coordinates. This approach allows precise execution of mobile device tasks based on visual analysis.

What is vision-first control for Android device automation?

Vision-first control for Android device automation is a technique where AI analyzes screen content to identify exact coordinates for interaction. It enables direct device manipulation by executing ADB commands for tapping, swiping, and typing based on visual interpretation rather than fixed coordinates.

Do I need ADB and a local vision model to control my Android device?

Yes, you need both ADB for executing device commands and a vision model for screen interpretation to control your Android device. ADB handles direct manipulation like tapping and typing, while the vision model analyzes screenshots to determine precise interaction targets.

Can I automate filling out forms within Android apps using screen analysis?

Yes, you can automate filling out forms within Android apps using screen analysis. The AI vision model interprets the screen content to locate form fields, and ADB commands execute the text input and tapping actions required to complete the forms automatically.

What are the limitations of using ADB commands with AI vision for mobile testing?

Using ADB commands with AI vision for mobile testing requires a connected device and depends on the vision model's accuracy for screen interpretation. Automation may be limited by dynamic UI changes, model processing speed, and the need for continuous ADB connection stability during task execution.