Screen text and touch control
Lets an agent work with visible iPhone content and perform core interactions such as tap, type, and swipe.
Open-source, self-hosted iPhone control for AI agents on macOS.
iphone-use is an open-source tool for giving AI agents control of a physical iPhone. It exposes screen text and interaction primitives—including tapping, typing, and swiping—through a Rust-based, self-hosted system intended to run on macOS. The project is designed for agents that need to observe a mobile interface and perform actions on the device rather than interact only with a simulator or static test fixture.
Lets an agent work with visible iPhone content and perform core interactions such as tap, type, and swipe.
Includes a custom XCTest-based runner as part of the device-control stack for driving iPhone automation.
Provides a command-line interface, HTTP API, and MCP integration so an agent workflow can use the interface that best fits its runtime.
Supports browser-oriented workflows in which an agent can take control of an existing mobile browsing session.
Supports recording or defining repeatable interaction flows that can be run again instead of reconstructing every action manually.
The project is implemented in Rust and is positioned for local operation on macOS, keeping the control service in the operator’s environment.
Build an agent that reads an iPhone screen, chooses an action, and executes taps, text entry, or swipes against a real device.
Turn a multi-step interaction into a replayable flow for recurring checks, demonstrations, or internal automation.
Let an agent continue work in a browser session that is already open on the iPhone, rather than starting from an empty context.
Expose phone actions through the CLI, HTTP API, or MCP so they can be connected to different agent runtimes and orchestration setups.
It is intended to control a real iPhone, with support for reading screen text and performing taps, typing, and swipes.
The project provides a CLI, an HTTP API, and MCP integration as control and integration surfaces.
The project describes itself as self-hosted on macOS. The supplied source does not provide a complete compatibility matrix for macOS or iPhone versions.
The repository is public and lists an MIT license. The source does not establish a separate hosted pricing plan for iphone-use.
www.cognee.ai
Cognee is an open-source memory platform for AI agents and applications. It connects data across sources, builds connected memory using vector, graph, and relational storage, and makes that context available through SDK, cloud, or self-hosted deployments.
steel.dev
Steel is open-source browser infrastructure for AI agents and web automation. It provides cloud browser sessions through an API, with support for common browser automation tools and workflows such as scraping, research, RPA, and agent-driven web tasks.
flyte.org
An open-source runtime for durable AI workflows, model training, serving, and agents.
superlinked.com
Open-source inference for agents, serving embeddings, rerankers, extractors, and open LLMs on your infrastructure.
theagentrouter.ai
Agent Router is an open-source Go project for managing unified access to generative AI services through Envoy Gateway. It is intended for teams that want a gateway-based control point for AI service traffic and related agent or MCP workloads.
busabase.com
An open-source database and workspace for connecting AI agents with reusable data, documents, skills, and apps.