iphone-use

The screen as text is what makes this work for me. an agent reading plain text is way more reliable than guessing from a screenshot

how does the remote control from the browser feel, is there much lag? also what's the 21-tool MCP server good at the most in Claude Code?

the part that gives me pause is the combo of "payment apps" and "remote control from any browser" in the same feature list. once there's an HTTP API and a browser panel that can tap through a real phone, that control channel is now the actual attack surface, not the apps themselves - whoever can reach that endpoint can do whatever the agent can do, including in banking apps. is there anything beyond "it needs a Mac with Xcode and the phone on USB" standing between that remote panel and someone who isn't you? local-only by default vs something you'd have to deliberately expose matters a lot here

Plenty of apps I use every day have no API: banking, payment, health, chat. I wanted my coding agent to be able to operate them on a real phone, the same way it operates a browser. iphone-use started as a remote-control panel and turned into an agent tool once it became clear what an agent needs that a person doesn't: an answer it can branch on (applied, not sent, unknown, retry-safe), refusals instead of taps that silently land elsewhere, and a way to replay a chore without spending tokens. Every request is timed, which is how tap-and-settle went from 8.5 s to 4.2 s. It needs a Mac with Xcode and the phone on USB. Feedback very welcome.