What a scraping browser is

A scraping browser is a headed browser process built to collect pages at machine pace. It is a driver with a window. It is not an operator profile. This definition exists so the two objects stop sharing a sales page.

The best-of commercial query lives under antidetect browsers for web scraping. That page skips the disguise. This page names the product people are actually buying when they say "scraping browser."

The job the phrase is supposed to name

A collector needs a process that can request a URL, run enough of the page to see the tree, and hand the tree to a parser. Bare HTTP misses JavaScript. A full Chrome with a person in the chair is slow. Vendors fill the gap with a headed (or headless) Chromium they control, often in a cloud, often with a proxy baked in, often with a stealth story.

That is a scraping browser. Persistence is optional. The cookie store is a fixture for the run, not a folder you reopen tomorrow for Ads Manager. The fingerprint, when they advertise one, is there to finish the harvest, not to keep Brand A apart from Brand B.

What web scraping is is the act. This page is one tool people use for the act. Browser automation versus a real profile is the architecture. Playwright, Puppeteer, and Selenium are the names you already know when you own the host. Playwright for pages you own.

How it differs from an operator profile

An operator profile is a directory you keep. Cookies, localStorage, a fingerprint seed, a proxy you brought. You sit in it. You close it. MaskWright 0.1 is that object on Windows. It is not a scraping browser. It has no collection API, no queue, no Playwright export.

A scraping browser is a process you throw at a list. The vendor may call the process a profile. If the session lives on their disk and the point is volume, it is still a collector. Browserless and Browserbase are not operator profiles on the automation desk. Same split.

We will not hide a scraper. Painting an operator fingerprint on a collector does not change the job. Anti-bot pages and official channels: a wall is a stop, not a reason to buy a prettier collector.

When a headed process is still honest

You own the host. The driver lives in your repo. The leftover is a test artifact. That is not a scraping-browser product. That is Playwright on furniture you control.

You are a person in a cold profile looking at a public page. That is not a scraping browser either. That is a profile with an empty cookie store. Instagram public research is look, not a harvest. Instagram public research versus a scrape. Public pages stay slow.

A cloud headed fleet against a site that forbade the method is the product we will not recommend. Personal data collection stays refused. Personal data and collection. Judge a tool on permission, not on render fidelity. How to judge a web scraping tool.

Proxies are not the definition

An exit can sit in front of a bare client or a headed one. A proxy is when the exit is enough. A residential line does not turn HTTP into a browser, and a browser does not become allowed because the exit changed.

Official Python libraries come first. Web scraping with Python, without a stealth stack. A scraping browser is what people reach for after they skipped that step.

If you cannot say whether the process will still be open tomorrow as a login you care about, it is a collector. If you can point at a folder you will reopen for Ads Manager, it is a profile. Buy the object you actually need. Do not rent a collector and call it a profile.

Local desk

If you need to see a public card, open a cold MaskWright profile and look. If you need volume, open the API docs. If you need to test your own site, open Playwright in a scripts folder. If a vendor demo starts with "undetectable headed Chrome at scale," you are looking at a scraping browser. We will not help you hide it, and we will not pretend it is a profile.

The Scraping hub keeps the definition here and the commercial no on the antidetect-for-scraping URL. Two pages, one line: a collector is not a profile.

FAQ

Is Playwright a scraping browser?

Playwright is a driver. On a host you own, it is a test tool. Pointed at a stranger's login at volume, it is doing the scraping-browser job, and that job is still a collector.

Is MaskWright a scraping browser?

No. 0.1 launches and isolates operator profiles. It has no collection API and no queue.