I
iphone-use
Let AI agents drive a real iPhone: the screen as text, tap/type/scroll with honest outcomes, over WebDriverAgent. HTTP API + MCP server, remote control from a browser or iOS app, replayable flows. Rust, self-hosted.
81 GitHub stars379 commitsMIT licenseRust codebaseMCP server with 23 tools
Opportunity score
54/100
Strong technical differentiation and rising AI-agent interest, but no visible monetization and a demanding self-hosted setup limit near-term business potential.
Founder verdict
MAYBE
A technically differentiated open-source iOS computer-use tool with real developer appeal, but no observed revenue model.
What is iphone-use?
iphone-use is an open-source, self-hosted Rust tool that lets AI agents and humans see and drive a real iPhone through a custom XCTest-based device runner compatible with WebDriverAgent. It exposes an HTTP API, an MCP server for Claude Code/Desktop, and a web/iOS remote control interface. The standout differentiation is honest action outcomes: every action reports whether it was applied, not sent, or unknown, and whether retry is safe. It also offers replayable 'flows' that run without a model, saving tokens for repeated tasks. Observed traction is early: 81 GitHub stars, 8 forks, 379 commits, MIT license, no visible pricing or revenue model.
Get the full iphone-use playbook
The complete reverse-engineering — what to copy, what to avoid, and exactly what to build instead.
- Full opportunity score across 8 dimensions
- The real problem & why customers pay
- Why it's winning — with evidence
- Complete business reverse-engineering
- Competitive advantages & moat analysis
- Weaknesses, risks & what to avoid
- Clone strategy — what to build instead
- Week-by-week MVP roadmap
- Recommended technical stack
- Founder verdict & highest-leverage move