More coverage. Less maintenance.

Q: Which platforms can AskUI automate?

AskUI works on any surface with a visible screen: Windows, macOS, Linux, Android, automotive HMI, industrial displays, and embedded interfaces like QNX. For devices where software installation isn't possible, AgentOS runs on a separate companion device. The target stays untouched.

Q: Do I need to write code to use AskUI?

No. Tests are written in plain English or imported from existing CSV/XLS test cases. Developers who prefer code can use the open-source AskUI SDK for Python.

Q: LLM-based means results could vary, how do you ensure consistency?

The more detailed your test description, the more consistent the output. Successful runs are cached as a replay file, so subsequent executions follow the same flow without calling the model again. When the UI changes, the agent automatically detects it and re-reasons from the new screen state.

Q: How does AskUI keep model costs under control?

Repeated runs become significantly cheaper over time as stable steps are cached and don't require model calls.

Q: Can AskUI run on-premises?

Yes. AskUI supports on-premises and air-gapped deployment. AskUI is ISO 27001 certified and GDPR compliant. Your data stays within your own infrastructure and is never used to train models.

Q: Can I try it before committing?

Yes. We offer a scoped PoC so you can validate AskUI against your actual use case before any long-term commitment. Contact us to get started.

For systems where state lives on the screen, not the DOM. Computer-use agents see it directly and verify deeper across APIs, backends, and devices.

Start trial Talk to us

macOS · Windows · Linux14-day free trialOn-prem available

Case studies

Real customers, real systems.

Production stories from teams using AskUI on POS devices, infotainment systems, industrial software, hospital workflows, and mobile business applications.

80% less testing time

Deutsche Bahn accelerates POS release testing with AskUI

GitLab pipeline automation for Android POS workflows at DB Fernverkehr.

Industrial control and manufacturing equipment

30 system tests

SEW Eurodrive scales system testing for industrial software

Azure DevOps pipelines across nested plugin workflows and desktop UI states.

Clinical workspace used for healthcare documentation

70% time saved

FICUS Health streamlines rehab documentation

Legacy hospital systems updated without custom API work or brittle selectors.

Mobile business workflow on a handheld device

130+ automated tests

Zucchetti replaces manual regression testing with AskUI

Canvas-based mobile and business software testing without fragile object IDs.

80%

Testing time saved

Deutsche Bahn integrated AskUI into GitLab pipelines for Android POS release validation.

AskUI cut our testing time by 80% and integrated seamlessly into our GitLab pipeline.

Umar Usman Khan, Quality Assurance Lead at DB Fernverkehr AG

Why it's hard

The more state, the more it breaks.

Most automation pain traces back to one thing: state. The more an interface carries, the more engineering it takes to keep scripts alive, until they stall at the maintenance wall.

Easy

Web

Mostly stateless, page state lives in the URL. Standard tooling copes.

Non-technical tester

Medium

Mobile

Frameworks exist, but device farms hurt and app state grows fast.

QA engineer

Hard

Desktop & HMI

State across the OS, files, IPC, and embedded screens. Brittle scripts.

Developer

Extreme

Distributed systems

State spread across services, queues, and caches. Needs full-system orchestration.

Senior engineer

Stateless

More stateful

AskUI's answer: a computer-use agent reasons about whatever is on the screen, so it absorbs state instead of breaking on it, the same plain-language workflow runs from a web form to a distributed system.

Agentic test automation

Write the test in plain English. The agent runs it.

A computer-use agent does what a tester does: sees the screen, reasons, acts. Use an existing test file or write instructions in plain text; it runs on any UI and reports audit-ready results.

From a spreadsheet, CSV, or sentence.

No selectors, no brittle scripts. The agent reads test steps like a tester would, reasons about what's on the screen, and executes them on the real interface.

Use existing CSV/XLS test cases
Author in plain text, no code
Runs on desktop, web, mobile, and embedded HMI
Structured, audit-ready report every run

The same agent will also document and operate your interfaces. Start with the test assets you have today, and turn those workflows into work instructions and operations next.

One platform

From pairing to a traced run.

Every test runs on the same runtime. Pair the device once, configure the agent, and watch every step.

Pair every device once.

Install AgentOS on macOS, Windows, or Linux targets. For embedded screens and devices you cannot instrument, use Companion Mode with HDMI capture.

Host Mode for full OS access
Companion Mode through a Raspberry Pi
CI/CD ready for VM logon and privilege flows

bench-04linux host mode

paired

head-unit-Aandroid companion

paired

kiosk-zone-2windows host

offline

pi-companion-2arm64 hdmi capture

paired

Configure prompts, tools, and models.

Each agent gets a system prompt, a model, and an allowlist of tools. Swap models without touching the prompt. Version everything.

Versioned system prompts
Per-agent model routing
Tool allowlists for controlled access

System prompt v4

Operate the target interface step by step. Capture a screenshot after each action, then report what you observed.

claude-sonnetcache on

Allowed toolsscreenshotcan-busfile-storehttp-fetch

Run it. Watch every step.

Trigger from the app, CI, a schedule, or an integration. Every action is traced, every screenshot is stored, every cache hit is visible.

Live run streaming
Cache hit and miss telemetry
HTML reports for audit and review

// run: head-unit-agent @ bench-04

[00.0s] screenshot   -> 1080x1920
[00.3s] locate       "settings gear"  [cache hit]
[00.4s] click        -> (912, 64)
[01.1s] screenshot   -> panel opened
[01.4s] locate       "Language"       [cache hit]
[01.6s] click        -> (240, 480)
[02.2s] scroll       -> "Deutsch" into view
[02.5s] click        -> tap dispatched
[03.1s] verify       label reads "Einstellungen"

done - 3.1s - 2 LLM calls - 4 cache hits

Ready to run your first agent?

14-day free trial · macOS, Windows, Linux

Start free trial Talk to us

Why teams pick AskUI

Computer-use is not the hard part. Production is.

Engineering leaders choose AskUI when they need agents to run reliably on real infrastructure.

Any device, any OS.

Windows, macOS, Linux, Android, iOS, plus embedded screens through Companion Mode when software access is not possible.

5+ OS targets

Any model, no lock-in.

Route each agent to Claude, Gemini, OpenAI, AskUI models, or your own provider through your credentials.

BYOM ready

Caching that pays for itself.

Skip redundant model calls on stable workflows and keep cost visible at every step of the run.

Cached steps cost $0

Built for production.

On-prem and air-gapped deployment, ISO 27001 certification, GDPR compliance, audit logs, and no training on customer data.

ISO 27001

FAQ

Common questions.

What is AskUI?

AskUI is an agentic test automation platform: you write tests in plain English and a computer-use agent executes them on desktop, web, mobile, and embedded HMI interfaces. Instead of querying the DOM or UI selectors, the agent observes the screen and reasons about what to do next, so it can test interfaces other tools cannot reach.

Which platforms can AskUI automate?

Do I need to write code to use AskUI?

LLM-based means results could vary, how do you ensure consistency?

How does AskUI keep model costs under control?

Can AskUI run on-premises?

Can I try it before committing?

For builders

Start trial.

Download AskUI Desktop, clone the demo project, or start with the SDK. Add API keys when you are ready to run agents.

Start trial

For teams

Ready for production?

Commercial AgentOS, bring-your-own-model, and on-prem for distributed fleets. We'll map a plan to your stack.

Talk to us

More coverage. Less maintenance.

Real customers, real systems.

Deutsche Bahn accelerates POS release testing with AskUI

SEW Eurodrive scales system testing for industrial software

FICUS Health streamlines rehab documentation

Zucchetti replaces manual regression testing with AskUI

Testing time saved

The more state, the more it breaks.

Web

Mobile

Desktop & HMI

Distributed systems

Write the test in plain English. The agent runs it.

From a spreadsheet, CSV, or sentence.

From pairing to a traced run.

Pair every device once.

Configure prompts, tools, and models.

Run it. Watch every step.

Ready to run your first agent?

Computer-use is not the hard part. Production is.

Any device, any OS.

Any model, no lock-in.

Caching that pays for itself.

Built for production.

Common questions.

Start trial.

Ready for production?

We value your privacy