Let an AI agent operate apps on an iPhone
An open-source developer project that uses GPT-4.1, Xcode UI tests and the accessibility tree to tap and swipe through iPhone apps.
Try it yourself
What you get
An agent that can carry out tasks in iPhone apps (developer project).
Tools
- PhoneAgent (open source)
- Xcode
- OpenAI API (GPT-4.1)
Steps
- Clone the repo and open it in Xcode.
- Add your OpenAI API key.
- Run it on a device and give it a task.
Proof / source
Open-source code Author's Show HN post and the GitHub repo.
Read the original source ↗ Discussion ↗
Related
Get calm, step-by-step coaching through a scam accusation call
Accused of taking $900, the user explained the situation to ChatGPT in voice mode and followed its guidance on a WhatsApp call; it turned out to be mistaken identity with…
Have a LangChain agent draw pictures in the ROS 2 turtlesim
A robotics grad student built a LangChain agent that controls ROS 2's turtlesim simulator to draw pictures from text instructions.