A browser automation agent is an AI system that operates a web browser for you, clicking, typing and reading pages to finish a task you describe in words.
In plain words
You give the agent a goal. It opens a browser, looks at the current page, and a language model picks the next action, such as click this link or type in that box. It repeats until the goal is met. Some tools show the model screenshots, while Microsoft's Playwright MCP server gives it the page's accessibility tree, a structured list of buttons and fields, so no vision model is needed.
A quick example
The Browser Use quick-start gives its agent the task of finding how many stars the browser-use repository has. When run, the agent drives a browser, a cloud one by default, and the program prints the agent's final answer.
Where you'll meet it
- Filling in forms and repetitive web tasks.
- Exploratory browser automation and self-healing tests.
Related
Sources
- Docsbrowser-use, Browser Use (GitHub) · read 13 Nov 2026
- Docsplaywright-mcp, Microsoft (GitHub) · read 9 Nov 2026
- Docsbrowser-use README, Browser Use (GitHub) · read 13 Nov 2026
Checked by WhatIsAI against the sources above · 13 Nov 2026. Spotted an error? Report it.