Does your resume sound like ChatGPT?

Recruiters now read hundreds of resumes that went through the same chatbots, and those resumes share habits: a small vocabulary of inflated verbs (spearheaded, leveraged, orchestrated), self-descriptions that claim a trait instead of showing it (results-driven, proven track record), bullets that stack "-ing" clauses ("Led X, driving Y, resulting in Z"), every line opening the same way, and every line the same length. We count those habits. The more of them your text has, the higher the score.

No, and neither can anyone else. We do not claim to detect who or what wrote a text. People wrote "spearheaded" long before chatbots did, and a chatbot can write plainly when asked. The score describes how the text reads to a tired recruiter, which is the thing that matters for your application. It is not evidence of anything about its author.

It means a reader may assume you did not. That is unfair, and it is also easy to fix: swap the flagged words for the plain ones, split the clause pile-ups into what you did and what happened, and let your lines be different lengths.

For up to three of your most flagged lines, we show a rewrite in plain language. The rewrite may only use what the line already says: it cannot add numbers, tools or claims, and we check each one before showing it. If a rewrite fails that check, we do not show it.

We don't keep your resume. The text is scored in memory and dropped when you get your result. Up to three lines are sent to our AI provider, Anthropic, to write the sample rewrites. If you choose to share your score, we store the number and the names of the flagged habits, never your text, for 30 days.

No. The check is free and needs no account. You can run it 5 times a day. The sample rewrites share a daily budget across all visitors, so late in the day you may get the score and the flagged lines without them.