ARTEMIS

ARTEMIS: Android automation tool that converts natural language instructions into device actions, enabling AI agents to control apps, run tests, and diagnose issues.

ARTEMIS allows users to automate Android workflows and testing via natural language commands. It connects with AI IDEs like Antigravity, Claude Code, and Windsurf through a native Model Context Protocol (MCP) server. When prompted, the tool executes tasks on connected physical devices or emulators, navigating screens and interacting with UI elements. It supports multiple execution profiles, including a Flash mode for routine tasks and a Pro mode for complex exploratory testing. The system utilizes accessibility hierarchies and visual models to target elements across standard and custom interfaces. It also generates diagnostic reports, including performance metrics and raw datasets, for further analysis.

ARTEMIS at a glance

Best for: Software engineer
Runs on: macOS, Linux, Windows
Pricing: Free
Released: August 2026
Category: AI coding tool
Website: github.com/google/artemis

Key features

Cross-app workflow execution on Android devices
Natural language instruction processing
Native Model Context Protocol (MCP) server support
Real-time device screen projection and interactivity
Pre-execution checks against live UI trees
Performance profiling and diagnostic reporting
Flash and Pro execution profiles
Support for physical Android devices and emulators

Use it for

Automating end-to-end mobile testing workflows
Running stability and exploratory tests on Android
Integrating mobile device control into AI coding assistants
Capturing UI logs and screenshots for debugging
Generating test reports with structured audit findings

Alternatives to ARTEMIS

Mobile MCP — Connects AI agents directly to real mobile devices for automated testing and tasks.
ARTEX — Runs autonomous multi-agent penetration testing with a dual-graph architecture.
agent-device — Token-efficient snapshots and semantic references for native mobile and TV apps.
QApilot — Explores mobile apps and self-heals broken tests without scripts.
TestMu AI — Autonomous AI agents and cloud execution deliver 70 percent faster test runs.
Midscene — Automates tests across web, mobile, and desktop using vision models.
ESP-Claw — Runs agent runtimes directly on device chips for local decision making.
CodeGraph — It eliminates file crawling, reducing AI agent tool calls by 88% and cutting costs.
JADX AI MCP — Connects JADX directly to LLMs for live reverse engineering and code review.
MeterSphere — Consolidates test management, API testing, and automation into one open-source platform.
Codeg — Run multiple AI coding agents together and manage them from mobile devices.
Oracle — Provides a single entry point for multiple pro AI agents and session management.

All ARTEMIS alternatives