Glossary
What is a browser agent?
A browser agent is an AI agent that operates a web browser: it reads pages, clicks, types, fills forms and navigates to finish a task. It may control a real browser through an extension or automation library, or see the screen and act like a user. In your own browser, logins stay with you.
How it controls the browser
| Method | How |
|---|---|
| Extension | Reads the page and acts from inside the user's browser |
| Automation library | Drives a browser via Playwright or Puppeteer |
| Computer use | Sees screenshots and moves the mouse and keyboard |
Local vs hosted
An agent in your own browser uses your existing sessions, so it's signed in where you're signed in, and sites see a normal visitor. A hosted headless browser on a server starts with no sessions and is more often blocked by bot protection. For submitting to directories, local usually works better.
Limits
Browser agents should stop at CAPTCHAs, two-factor prompts, payments and terms acceptance. They should also treat page content as data, not instructions, to resist prompt injection.
Example
A browser agent in a founder's Chrome opens 30 directory submit pages in turn. It fills each form from her product details, uploads her logo, waits for her at two CAPTCHAs and records each listing's status.
FAQ
Can a browser agent log in for me?
It can use sessions you're already signed into. Entering passwords should stay with you or a password manager.
Do websites block browser agents?
Many block automated traffic, especially from servers. Agents in your own browser are blocked less.
Is a browser agent the same as a scraper?
No. Scrapers extract data; browser agents complete tasks.