ATP360 Getting Started

Introduction What ATP360 does

ATP360 is WorkspaceDNA’s agentic application testing and validation platform. It turns documents, videos and prompts into structured, repeatable test cases, executes them on real Windows devices using a computer-using agent, and records fully auditable results — every action taken, the models used, the tokens consumed and the cost of each run.

The platform is built for user acceptance testing and build validation of Windows applications, where the same journeys need to be re-run against new builds, new operating system versions and new patches, with a human able to check and correct the AI’s verdict at any point.

The three-layer AI engine

Every test run passes through three layers. Each layer can use a different AI engine and model, and each is configurable both per run and as a workspace default.

Before you start

If you are standing the platform up yourself — particularly for a PaaS deployment — read the Architecture, Security and Deployment Pre-Requisites guide on the knowledge base first. It covers the moving parts and what must be in place before provisioning. Then work through the ATP360 Deployment Guide, which walks you through setting up the PaaS solution step by step.

This document assumes ATP360 has already been provisioned for you and that:

• your workspace exists and you can sign in;
• at least one AI engine is connected under Settings > AI Integrations;
• at least one CUA worker is installed and online;
• any credentials your applications need are stored as secrets.

Workers, engine connections and secrets are all covered in the ATP360 Administration Guide.

Finding your way around

The navigation bar runs down the left of every screen. The upper half holds the working areas:

• Home — the dashboard, and the quickest way to drill into recent results.
• Test Now — compile and run an on-demand test.
• Test Results — every run that has completed.
• Test Cases — your library of reusable test cases.
• AI Efficacy — human review of AI verdicts, and the overrides you have made.
• Pass/Fail Status — passed and failed runs over time.
• Test Failures by Type — why runs failed: Stopped Manually, Step Limit Reached, Criteria Not Met, AI Could Not Complete, or Other.
• Overridden Results Over Time — the override rate, i.e. how often a human disagreed with the AI.
• Executed by OS — runs broken down by the operating system of the worker they ran on.

The lower half holds Feedback, Settings, Support and Logout. Directly beneath the WorkspaceDNA logo is the workspace selector, which appears only if you have access to more than one workspace.

The Home dashboard

Home is a high-level dashboard of what ATP360 has been doing. The period it covers is set by the range picker in the top right (for example Last 7 Days or Last 30 Days).

Figure 1 — The Home dashboard. The tiles across the top summarise the selected period; the charts below break the same data down by outcome, failure type and operating system.

Four charts sit below the tiles:

Figure 2 — Each chart carries its own drill-down hint — “Click a bar to view its tests”, “Click a slice to view its tests”, “Click a point to view its tests”.

Everything on this screen is a way into the detail. If you want to see the test cases that failed because their criteria were not met, click that segment of the Test Failures by Type chart: you get the matching runs with their state, action count and cost, and the view icon opens any of them in full.

Figure 3 — Drilling into a chart segment — here Criteria Not Met — lists the matching runs, each of which can be opened in full with the view icon.

This is the short path from nothing to a signed-off result. Every step has a fuller treatment in ATP360 Running Tests; the point here is to see the whole loop once.

Figure 4 — The Test Now screen, where a test is compiled and run: the AI engine pipeline, the worker to run on, and the test source.

1. Open Test Now. Leave the AI Engine Pipeline panel as it is — the workspace defaults are already set.
2. In the CUA Worker panel, pick a worker with a green dot. Green means online and ready; a red dot cannot be selected.
3. Under Provide test source, choose From prompt and describe the test in plain English — for example “Open Notepad, type hello then load the ATP360 Worker screen which will be open in the background”. Be specific about what should happen.
4. Click Extract test steps. ATP360 turns your description into numbered steps, each with an action and an expected outcome. Read them before going further.
5. Run the test. The live view shows the test steps on the left, the worker’s screen in the middle and the AI’s action log on the right, with duration, tokens and cost ticking up along the top.
6. When it finishes, the result appears at the bottom and asks you to confirm it: Approve if the AI got it right, or Override — with a reason and notes — if it did not.

Figure 5 — The end of a run: the AI’s verdict, with Approve and Override for the human checking it.

Workspaces, feedback and support Switching workspace

If you have access to more than one workspace, a selector appears beneath the WorkspaceDNA logo showing the one you are working in; click it to switch. Most people have a single workspace and will never see this control — if it is there, it means you have several.

Figure 6 — The workspace selector, listing the workspaces you can switch between.

Feedback

The Feedback button in the lower half of the navigation opens an external feedback board. Use it for anything you would like to see in ATP360, or for general feedback — it is not the route for support issues. 

Support and the knowledge base

Support opens the WorkspaceDNA knowledge base. The ATP360 section holds the guides that sit alongside this set, including the architecture and security guide (how the platform is put together, what the moving parts are and what must be in place before provisioning), the deployment guide, known and fixed issues, and an explanation of the test failure types.

Figure 7 — The ATP360 area of the WorkspaceDNA knowledge base.

Getting help

If something in ATP360 is not covered here, start with the knowledge base — reachable from Support in the navigation — which carries the deployment and architecture guides, the list of known and fixed issues and an explanation of the test failure types. The other documents in this set are listed on the title page.

If you still cannot find the answer, contact WorkspaceDNA supporte on technicalsupport@workspacedna.com.

Glossary