Real work in real apps — with the lights on
App Use applies to almost any application with a user interface. Wherever an agent would otherwise click blindly or shell out to a black box, App Use makes the work observable and governable.
Agent-driven testing & QA
Let an agent exercise your desktop app the way a user would — fill forms, invoke actions, read back state — and get a reproducible, step-by-step record of everything it did. Because the agent reads the app's declared surface, tests don't shatter the moment a button moves a few pixels.
Transparent app automation
Automate repetitive workflows across the apps your team already uses. Unlike a script that silently mutates files, every step runs through the app's real UI state and is metered — so you always know which app and which feature the agent operated.
Enterprise governance & compliance
Give auditors a tamper-evident trail of every agent action, scope each agent to read-only or read-write, and require a human to approve sensitive operations. Stream the audit chain to your SIEM. Identity flows from Microsoft Entra ID across every federated hub.
Cross-device orchestration
One agent, one connection — driving apps on a desktop, a laptop, a phone, and a server appliance as if they were on one machine. Peers cluster over mutually-authenticated TLS on the LAN, or an end-to-end-encrypted relay across the internet.
High-stakes operations with a human gate
For actions that move money, change configuration, or touch production, the consent gate stops the agent and asks a person first — showing the exact action, in a GUI, before anything happens.
Beyond the desktop
App Use is transport-neutral: the surface model is the same whether an app is desktop, server, or browser-based. Extending the surface to web pages is an active area of work — tell us your scenario.
Have a workflow in mind?
Describe what you want an agent to do, and in which apps. We'll tell you how App Use makes it transparent and safe.
Talk to us