Browser agent
A browser agent is an AI agent that controls a web browser to complete tasks, such as searching, clicking, filling in forms and reading pages, often running a headless browser with no visible window.
Also called: headless browser, browser automation
By Divya Sharma · Checked Nov 15, 2026 · How we review
Much of the web has no API. A browser agent works through the same pages a person uses.
How it works
The agent drives a real browser through an automation library such as Playwright, which can run Chromium, Firefox and WebKit, headless by default. On each step the agent reads the page, either as a simplified version of its structure or as a screenshot, then asks for an action: go to a URL, click an element, type text. The automation layer performs it and returns the new page state.
A headless browser is the browser running without a window. It is the plumbing, not the agent; scrapers and test suites use it too. Tools for agents include the Playwright MCP server, Browser Use and Stagehand, plus hosted browser services such as Browserbase.
How it differs from related terms
Web crawling collects pages at scale and does not interact with them. Computer use controls a whole desktop through screenshots, not just a browser.
Limits
Web tasks remain hard. On the WebArena benchmark of realistic sites, the best GPT-4-based agent in the original paper completed 14.41% of tasks, against 78.24% for humans. Logins, captchas and changing layouts break agents. Every page is untrusted input, so a browser agent is a common target for prompt injection.
Related terms
- Computer use
- Computer use is the ability of an AI model to operate a computer the way a person does, by looking at screenshots and sending mouse clicks and keystrokes, so it can work with software that has no API.
- Web crawling for AI
- Web crawling for AI is the automated fetching of web pages, by following links or a list of URLs, to collect text for training models, building search indexes or giving an application fresh content to retrieve.
- AI agent
- An AI agent is a system in which a language model decides its own next steps: it calls tools, reads the results and repeats until the task is done, instead of following a fixed script.
- BrowseComp
- BrowseComp is an OpenAI benchmark of 1,266 hard questions that tests whether a browsing agent can search the web persistently for tangled, hard-to-find facts, with short answers that are easy to check.
- Prompt injection
- Prompt injection is an attack where text fed to a language model, typed by a user or hidden in a web page, file or tool result, changes the model's behaviour in ways its developer did not intend.