Most websites don’t know what their own pages send. Marketing adds a pixel, a plugin adds another, the tag manager adds three more, and the cookie banner makes everyone feel it’s handled. Then a tag fires before the visitor answers the banner, or keeps firing after they click reject, and an email address ends up at an ad network.
PXTA, short for PiXel Tracking Audit, checks. It crawls a site in a real browser three times, once with the banner left alone, once after reject and once after accept, records every request each tracker sends, and points to the personal data inside. Every finding is a plain sentence with the request that proves it. The site calls it evidence, not a score.
The Pitch in Four Numbers
- 3 passes per page: banner unanswered, after reject, after accept, each in a fresh browser profile.
- 10 trackers recognised by name, from Meta, Google and TikTok to LinkedIn, Microsoft, Pinterest, Snap and X, plus unknown beacons.
- 500 pages in the endurance test. The browser was killed at page 167, PXTA restarted it, and all 500 were recorded.
- Apache 2.0, free, with no telemetry. Reports stay on your machine.
How It Works
PXTA drives Chrome, Chromium or Edge, so pages run their scripts exactly as they would for a visitor. It records requests from the page, its iframes and its workers, plus the beacons sent as the visitor leaves. Then it decodes what’s inside, URL encoding, base64, JSON and nested addresses, and looks for emails and their hashes, phone numbers, names, user IDs, search terms, and visits to health or financial pages.
The output is one self-contained HTML report, the same findings as JSON, and a HAR file whose checksum the report prints, so the evidence can’t be quietly swapped. A finding reads like a sentence a lawyer can use: this pixel sent this email address in the page address before consent.
Version 0.2 added a check that fails a build when a tag fires before consent, so a bad change can be stopped before it goes live. Version 0.3 compares two audits and shows what a tag manager change added.
What It Finds
The site’s examples, built on made-up businesses from PXTA’s test fixtures, show the kind of thing that turns up. A clinic’s pages sent data to TikTok and LinkedIn while Google Analytics politely waited for consent. A lender’s reject button loaded ad tags anyway. An order confirmation page sent a customer’s email to Meta, Google and Pinterest before consent. A newsletter form sent the hash of an email nobody had submitted. And a post explains why a hashed email still identifies someone.
Who It’s For
- Privacy and legal teams running a quarterly tracking sweep.
- Agencies taking over a client site, who want to know what they’re inheriting before it becomes their problem.
- Web teams who want a build to fail when a tag jumps the consent banner.
- Health, finance and other sensitive sites, where the stakes of a leaky pixel are highest.
Limits
PXTA doesn’t decide what’s legal. The report says plainly that’s a lawyer’s call. It can’t see server-side data such as conversion APIs or server-side tag managers, because that never passes through the browser. It doesn’t sign in, submit forms or get past anything a visitor couldn’t. It crawls politely, one page at a time with a pause, follows robots.txt and names itself in the user agent.
The Four-Question Screening
Every project featured here gets the same four questions, the way you’d screen a startup idea.
1. Where is the money in it?
PXTA is free and has no paid tier. The money is in the risk it measures. In the US there are class actions over pixel use, health regulators have warned hospitals and clinics about trackers, and European fines for consent failures keep coming. Companies pay to know where they stand. The paid paths are scheduled monitoring across many sites with alerts, reports formatted for legal and audit teams, an agency edition for handovers, and expert review on top of the evidence for companies that want a person to read it with them.
2. What real problem does it solve?
The gap between what a site is supposed to send and what it actually sends. Teams usually learn what their site sends from marketing’s description of it, which is often incomplete. Cookie banners give a feeling of control that the network traffic doesn’t always back up.
3. The technology in plain words
Imagine visiting the same shop three times: once walking in and saying nothing, once saying “please don’t track me,” once saying “go ahead.” Each time someone stands at the door and writes down every note the shop passes out the back, and what’s written on it. PXTA is that person at the door, and it hands you the notes.
4. Moat and differentiation
Cookie scanners list cookies and sometimes the scripts that set them. PXTA looks at what leaves the browser, the content of each request and its timing against the consent choice. A site can set no new cookie and still send an email address to an ad network, and only the second approach catches it. The moat that can grow is the tracker and banner catalog, kept on GitHub as JSON, and the decoding that finds personal data hidden inside requests.
Why It Matters
Privacy enforcement has moved from cookies to what actually gets sent. Courts and regulators want evidence, and so do the teams defending a site. A free tool that produces it, request by request, makes the conversation between marketing and legal a lot shorter.
Follow the Project
Start with the quick start, look at the sample reports, get a build from the download page, or read the code in the PXTA repository on GitHub. PXTA is listed with the other open-source projects on SoftDevStudio.com, and more featured projects are in the Projects section.