AltTester® CLI: Built for the way AI-assisted testing actually works

We’re excited to announce the release of the AltTester^® CLI – shipping as part of the updated version of AltTester^® Desktop (v2.3.2).

Game testing workflows are changing. More teams are integrating AI coding assistants and agents into their development pipelines, and the tools those agents rely on need to be scriptable, composable, and operable without a graphical interface. The AltTester® CLI is built for exactly that: plugging AltTester^®‘s capabilities into whichever AI tool or automation layer your team already uses, whether that’s Claude Code, Copilot, another coding assistant, a CI/CD pipeline, or a custom agentic workflow. You can use it directly from the terminal too, but the design is optimised for integration, so your AI tools can handle the testing while your team stays focused on the build.

If you’re building AI-assisted testing workflows, integrating AltTester^® into an agentic pipeline, or looking to give your AI tools direct access to game object inspection, input simulation, and state verification, this release is for you.

Why a CLI for game testing?

Many game testing tools are built around graphical interfaces. That works well for exploratory testing, but it’s a bottleneck the moment you want AI agents or automated pipelines to participate in the testing loop.

A command-line interface (CLI) changes the picture entirely. Every capability the AltTester® SDK exposes – finding objects, simulating inputs, reading game state, managing devices – becomes a command that an AI agent can call, chain, and reason about within its own workflow. The agent doesn’t need to understand a GUI. If it gets a structured interface, it can work with it directly.

Introducing the AltTester^® CLI

What is AltTester^® CLI?

The AltTester® CLI is a new component shipping with AltTester® Desktop 2.3.2. It connects to any Unity or Unreal game instrumented with the AltTester® SDK and exposes game interaction (object discovery, input simulation, state inspection, and test execution) as a set of clean, scriptable commands that AI agents and automated pipelines can call directly.

How it connects:

• Runs as a persistent background daemon over WebSockets (including secure wss://)
• Connects to port 13000 by default

Supported platforms:

• macOS — Apple Silicon & Intel
• Windows — x64 executable
• Linux — x64 & ARM64 binaries

Find any object, any way you need

The CLI gives AI agents a comprehensive set of object discovery strategies, so any automated workflow can locate game objects precisely.

Search by:

• Name — find objects directly by their name in the scene
• Hierarchy Path (XPath) — navigate the full scene tree
• Instance ID — target a specific object instance
• Tag — filter by Unity or Unreal tags
• Render Layer — scope your search to specific rendering layers
• Text content — find objects by the text they display
• Component type — locate objects by the components attached to them
• Screen coordinates — identify elements at a specific X/Y position on screen

Query modifiers:

• contains — partial string matching for more flexible searches
• all — return all matching elements, not just the first
• wait + timeout — poll until an element appears, ideal for async or loading states

Simulate any input – mouse, keyboard, touch, and more

Realistic input simulation is at the heart of game testing, and the CLI covers all the bases across desktop and mobile scenarios. Every input type is a command that an agent can issue programmatically.

Desktop inputs:

• Mouse movement, clicking, and scrolling
• Granular keyboard controls: press, key-down, and key-up

Mobile inputs:

• Taps and swipes
• Multi-touch gestures with finger-id tracking, accurately modelling begin, move, and end states across multiple contact points

UI controls:

• Read and inject text directly into UI input fields

Inspect and manipulate game state at runtime

This is where AI-assisted testing becomes most useful. The CLI gives agents deep visibility into – and control over – the game’s internal state while it’s running, enabling verification beyond “did it crash.”

Game state:

• Export the full active scene object hierarchy as JSON using game-state or snapshot
• Output inline or save directly to a file

Property reflection:

• Read or overwrite specific component properties at runtime by referencing the component name and its assembly

Environment control:

• Manipulate the game engine’s time scale — pause, slow down, or fast-forward game time for timing-sensitive test scenarios

Media:

• Capture screenshots alongside your test runs
• Query screen resolution

Mobile device management – Android and iOS

The CLI also wraps native device management tooling for teams testing on real Android and iOS devices, making it scriptable from the same interface. Full details on supported commands, platform requirements, and setup are in the documentation.

Project setup and test execution

Robot Framework integration:

• Run .robot test files or individual keywords directly from the command line
• report.html and log.html output generated automatically

Project scaffolding:

• Generate boilerplate code for new test automation projects in Python or .NET
• Spend less time on setup and more time writing tests that matter

AltTester® Skills for AI coding assistants — now in Beta

Running alttester install-skills installs an AltTester® skill into your AI coding assistant, giving it the context it needs to work with AltTester® effectively, without repeating context every session. It is compatible with Claude Code, Copilot, and other AI tools.

When you run the command, you’ll be prompted to choose between two skills:

1. Robot — designed for teams that don’t have a test framework yet and prefer not to write test code manually. The AI will guide you through planning, generating, running, and recording tests using Robot Framework’s keyword-driven syntax, which reads closer to natural language than traditional code.

2. Helper — designed for teams that already have a test framework and existing tests in place. Works across languages (C#, Python, Java) and helps with debugging failures, fixing broken tests, and writing new tests that fit your existing project structure.

Once installed, the skill is available directly in your AI assistant, so you work with AltTester® through the tool you’re already using, in whichever mode fits your project.

It’s currently in beta. Try it and share what’s working and what isn’t — support@alttester.com or Discord.

Demo: see AltTester^® CLI in action

The best way to understand how the CLI fits into an automated workflow is to see it running against a real game build. The demo below walks through connecting to a live Unity game, discovering objects in the scene, simulating inputs, and inspecting game state, all without a graphical interface.

📽️ Watch the demo below 👇

---

Explore these updates with a free 30-day Pro trial

The AltTester^® CLI works as part of the broader AltTester^® Pro ecosystem – our full-featured plan for professional QA teams and studios.

If you haven’t tried Pro yet, now is a great time to start:

✅ 30 days free, no credit card required

✅ Access to the full suite of AltTester^® tools and features

✅ Set up your project, connect the CLI, and see a complete game test automation workflow in action

👉 Start your 30-day free trial.

---

Get started

Explore the AltTester^® CLI feature as part of AltTester^® Desktop 2.3.2 and check the full documentation for platform-specific installation instructions and examples.

👉 Download AltTester® Desktop 2.3.2. 

Have questions or want to share feedback? We’re on Discord and always happy to hear from you – drop us a message or reach out at contact@alttester.com.