Paste the email, report, or post you are not sure about and get a clear detector result in seconds, before it goes out under your name.
3 free checks per day
Its default style has a handful of tics. None on its own is proof, but stack a few together and a draft starts to read machine-made. Claude also began marking its own output in August 2026, which answers a narrower question than most people assume.
It hedges both sides before landing anywhere
On one hand this saves time, though it is worth noting there are tradeoffs to consider as well.
It closes by offering to keep going
Let me know if you would like me to expand on any of this further.
It reaches for structure even in plain prose
There are a few key factors to keep in mind: first, second, and finally.
The score reflects how machine-written the text reads overall, not which model produced it. Claude's default style, careful hedging, offers to continue, reaching for structure, is the framing here, but the same check works on text from any model.
Rebuild the passage around your own structure, vary sentence rhythm naturally, cut hedging, and rescan. A light polish on Claude's structure rarely changes how the passage reads.
Yes. This page runs the same free, no-account check: 150-500 words a scan, 3 scans a day. Sign up for the full sentence-by-sentence report and a higher monthly word allowance.
See what gets flagged, then revise, rescan, and compare every version.
Start free