Chrome extension · v1.0

Scan for AI patterns anywhere you write

The GPTypo false-positive shield, now in your browser. Select text on any page, scan it against the same patterns detectors look for, and fix flagged sentences in place — without leaving Gmail, your CMS, or the form you're filling out.

Free · no account required to scan · Chrome Web Store

The GPTypo extension side panel scoring a LinkedIn post: a 62 Mixed verified score, flagged sentence with AI vocabulary and generic-phrasing tags, and a suggested rewrite with a Replace button.
01Where it works

The extension runs on the surfaces writers actually use. On writeable fields you can Replace a flagged sentence with a suggestion in place. On read-only pages you get Copy.

Gmail
Scan + Replace
LinkedIn
Scan + Replace
Substack
Scan + Replace
Notion
Scan + Replace
Reddit
Scan + Copy
Anywhere else
Scan + Copy
02How it works

Same short loop as the web scanner — scan, review, fix — without the copy-paste round trip.

1
Select and scan
Highlight text on any page and click the floating Scan button, use the right-click menu, or open the side panel. You get a 0–100 score in seconds.
2
See what trips detectors
Each sentence is scored on its own, with plain-language reasons for every flag — uniform lengths, AI-tell phrases, generic phrasing, and more.
3
Fix it in place
Click a flagged sentence for a tone-matched rewrite. Replace it inline on writeable fields, or copy it out on read-only pages.
03Free vs Pro

The extension is free to install and use. Free runs the fast heuristic scanner (500 words a scan, 5,000 a month). Pro unlocks verified neural scoring, unlimited rewrites, and shareable proof reports — the same account works on the web and in the extension.

04Privacy

The extension never reads a page on its own. It only acts on text you explicitly select and scan. That text is sent to the GPTypo API for scoring and is never sold. Read the privacy policy →

Read next
How GPTypo works → the scan, review, fix loop in full

GPTypo is a false-positive shield for human writers, not a tool for disguising AI text. We publish our accuracy data openly.