Mirroir FAQs

Question 1

What is Mirroir?

Accepted Answer

Mirroir enables remote control of iPhones and macOS windows, allowing users to capture screenshots, tap, swipe, type, and more. It's built on the Model Context Protocol (MCP) for versatile UI automation and testing.

Question 2

How does Mirroir enhance UI automation and testing?

Accepted Answer

It features AI-powered Agent Diagnosis for automated failure analysis, YAML-based scenarios for descriptive test flows, an integrated Test Runner for CI, and a Recorder to generate scenario YAML from user interactions.

Question 3

What platforms can Mirroir automate?

Accepted Answer

Mirroir can control iPhones through macOS iPhone Mirroring and any macOS window, requiring macOS 15+. It provides powerful automation capabilities without needing access to application source code.

Question 4

How does Mirroir diagnose automation failures?

Accepted Answer

Its Agent Diagnosis feature identifies *why* a compiled scenario fails using a two-tier system: deterministic OCR analysis and optional AI model integration (e.g., Claude, GPT-4o), providing root cause analysis and suggesting YAML edits to fix issues.

Question 5

Is Mirroir compatible with popular AI development clients?

Accepted Answer

Yes, Mirroir functions as an MCP server and seamlessly integrates with various MCP clients like Claude Code, GitHub Copilot, Cursor, and OpenAI Codex, allowing you to control devices directly from your AI agent.

Mirroir

Mirroir

Key Features

Use Cases

Key Features

Use Cases