software-mansion/argent

An agentic toolkit to control, debug, and profile iOS and Android apps. Made by Software Mansion.

What it solves

Argent is an agentic toolkit designed to give AI assistants direct control over various development environments. It eliminates the need for developers to manually switch contexts between their code editor and device simulators, allowing an AI agent to handle repetitive tasks like reproducing bugs, testing features, and profiling applications across multiple platforms.

How it works

Argent acts as a bridge between an AI agent (via the Model Context Protocol - MCP) and the target platforms. It provides a set of tools that the agent can use to interact with the UI, manage devices, and perform diagnostics. It supports a wide range of targets including iOS Simulators, Android Emulators/physical devices, Apple TV, Android TV, Fire TV, and Electron/Chromium apps.

Who it’s for

Developers who use AI coding assistants (such as Claude Code, Cursor, or VS Code) and want to automate the UI interaction, testing, and debugging of mobile, TV, and desktop applications.

Highlights

  • Multi-platform support: Control iOS, Android, TV (Apple TV, Android TV, Fire TV), and Electron/Chromium web apps through a single toolkit.
  • Autonomous UI interaction: Enables agents to perform taps, swipes, pinches, typing, and mouse/keyboard actions exactly as a user would.
  • Deep diagnostics: Ability to inspect logs, capture network traffic, evaluate JS, and walk component trees (UIKit and React).
  • Profiling tools: Integrated support for React Native (Hermes), React DevTools, and native profiling (Xcode Instruments/Android Perfetto).
  • Visual regression: OCR- and font-aware screenshot comparison to detect UI changes.
  • React Native first: Works out-of-the-box with React Native projects without extra setup.

Related

  • Project
    callstack/agent-device

    A tool that gives AI coding agents a live feedback loop to inspect, control, and verify mobile, TV, and desktop apps across iOS, Android, and other platforms.

  • Project
    agnt-gg/agnt

    AGNT is a self‑hosted desktop/Docker app that acts as a local‑first operating system for AI agents. It combines persistent agents, a visual DAG workflow canvas, long‑running goals, memory/insights, a plugin marketplace and a local HTTP + Socket.IO API, all configurable with dozens of LLM providers. Use it to automate events, run complex multi‑agent projects, keep data on‑premise, and turn execution traces into reusable skills.

  • Project
    AAswordman/Operit

    An open-source AI Agent platform for Android that integrates LLMs with system automation, a built-in Ubuntu terminal, and project workspaces for mobile productivity.

  • Project
    macOS26/Agent

    A native macOS application that provides an autonomous agentic AI capable of executing tasks, coding, and controlling Mac apps and browsers directly on the desktop.

  • Project
    coinbase/agentkit

    A toolkit for giving AI agents crypto wallets and onchain interaction capabilities, enabling them to to handle payments and monetize their services.