Skip to main content

CAPTCHA guide

Lemin Captcha

Image puzzle where you pick the crop that completes the picture.

Difficulty: Medium
Image puzzle
Exploded concept drawing of how the Lemin challenge is built

Lemin Captcha, served from leminnow.com, is an image puzzle. Its best-known variant is "Cropped": the visitor chooses which crop fills the gap in the main picture. There is no behavioural engine behind it and the JavaScript API is tidy, so it is one of the more approachable CAPTCHAs to automate.

How to identifyLink to How to identify

  • A container <div>, often id="lemin-cropped-captcha"
  • The script comes from api.leminnow.com, with the captchaId (CROPPED_...) in its URL
  • A global window.leminCroppedCaptcha exposing getCaptcha(), getCaptchaValue(), isReady() and reloadPuzzle()
  • lemin_challenge_id and lemin_answer appear in the form when it is submitted
  • Server-side checks go to https://api.leminnow.com/captcha/v1/cropped/validate

How it worksLink to How it works

  1. The page boots the widget with its captchaId and the divId of its container.
  2. Lemin answers with an image puzzle and a challenge_id.
  3. Once the visitor picks the right crop, the widget keeps an encrypted answer.
  4. Submitting the form sends lemin_challenge_id and lemin_answer along.
  5. The site's backend posts {private_key, challenge_id, answer} as JSON to /captcha/v1/cropped/validate and gets back true or false with an error code.

Free bypass methodsLink to Free bypass methods

Method What it does Success Rate Link
OpenCV Scores each candidate crop against the gap using edges and colour 75-90% opencv/opencv-python
DrissionPage Drives a real browser, then reads getCaptchaValue() 80-95% g1879/DrissionPage
Playwright Clicks the answer and pulls out {answer, challenge_id} 80-95% microsoft/playwright-python
ddddocr Fallback classifier for the crops 60-80% sml2h3/ddddocr
Patchright Stealth Playwright for sites that fingerprint the browser 70-90% Kaliiiiiiiiii-Vinyzu/patchright-python

Practical approachLink to Practical approach

  1. Pull the captchaId out of the script URL and note the container's divId.
  2. Load the page in a real browser and wait until leminCroppedCaptcha.getCaptcha().isReady() is true.
  3. Capture the main image and every candidate, then rank candidates by how well their edges and colours continue the gap.
  4. Click the highest-ranked crop.
  5. Call getCaptchaValue() to get {answer, challenge_id} and submit straight away.

Practical tipsLink to Practical tips

  • The widget encrypts the answer, so forging it is a dead end: always read it back with getCaptchaValue().
  • A challenge_id belongs to one widget instance. Reloading after solving throws the answer away.
  • With several widgets on a page, target one explicitly via window.leminCroppedCaptcha.getCaptcha('CAPTCHA_KEY').
  • Lemin looks at network traffic too: solve and submit through the same IP.
  • When your matcher is unsure, reloadPuzzle() gives you a fresh puzzle. A few retries usually beat a bad guess.

Skip the solver: let BrowserSolver handle Lemin

Website AntiBot runs a real browser session with matching TLS, headers, cookies and IP, clears the challenge, and returns the page. You send a URL, you get HTML back.

For education and authorized testing only. Respect the law, privacy, and the terms of the sites you work with. Difficulty labels are editorial estimates. Adapted from the CC0 community field guide on GitHub.