Getting Started With AskUI

TLDR

AskUI helps you automate real user interfaces by operating at the OS level and using visual understanding instead of DOM selectors. This makes it useful for UI automation across apps and environments where selectors are missing, unstable, or not available.

If your target is a web-only app with stable selectors, a DOM-based tool like Playwright may be simpler. If you need cross-app or OS-level UI automation, AskUI is a better fit.

What is AskUI?

AskUI is an Agentic Infrastructure designed for seamless, OS-level interaction. Unlike tools limited to the browser's DOM, AskUI's agent observes the screen visually to execute clicks, typing, and navigation across any application. It automates exactly what a human user operates at the OS layer.

When to use AskUI vs DOM-based tools

Use a DOM-based tool like Playwright, when:

• You automate a web-only application
• Selectors are stable and accessible
• Your workflow stays inside a single browser context

Use AskUI when:

• You need cross-app workflows (browser + desktop app, multiple apps, OS dialogs, etc.)
• Selectors are missing or unstable
• The UI is custom-rendered such as canvas-like surfaces or otherwise hard to target reliably

2026 Setup Checklist

To ensure your AI agent has the "Eyes" and "Hands" it needs, verify these requirements:

• Workspace Credentials: Obtain a valid UUID-formatted Workspace ID and Access Token from the AskUI Hub
• Local Runtime: Install the AskUI Suite (Agent OS) on your execution environment for local mouse and keyboard control.
• Python Environment: Use Python 3.10+ and install the core package: pip install "askui[all]".

Choose Your Starting Point

Instead of repeating basic tutorials, use this guide to jump into the specific path that fits your project:

• Path A: Python SDK Quickstart

  Best for a direct, “code-first” approach. Focuses on the first run, screenshot aritifacts and basic tool usage

  Getting Started: Computer-Use Agents with the AskUI Python SDK

• Path B: End-to-End Demo Project (Advanced)

  Best for developers needing a production-ready structure. Features CSV-driven steps, HTML reporting, and the AskUI Tool Store.

  AskUI Demo Project on GitHub

• Path C: Architecture Deep Dive

  Best for understanding the “Why” before the “How.” Explains the separation between perception and execution

  Understanding AskUI: The Eyes and Hands of AI Agents

FAQ

Q: Is it possible to automate multiple apps in one workflow?

A: Absolutely. AskUI's OS-level operation allows it to transition between browsers, desktop apps, and system settings seamlessly.

Q: Can AskUI handle apps without HTML tags?

A: Yes. By using visual intelligence, AskUI identifies elements on screen regardless of the underlying code or DOM structure.