reCAPTCHA v3 solver

reCAPTCHA v3 never shows a checkbox. It scores the visitor silently in the background and hands the page a token, so there is nothing to click and nothing to read from an image. Our API produces that token for you: send the sitekey and the page URL, then submit the token back into the page.

What reCAPTCHA v3 actually is

Version 3 is a scoring system, not a challenge. The script runs on every page and watches behaviour, then asks Google for a token together with a score between 0.0 and 1.0. The site decides where to draw the line. Because there is no widget, there is also no image to solve and no checkbox to tick, which is why every v3 solve on the market is a token solve rather than a recognition solve.

Why v3 is harder to automate than v2

There is no visible widget, so nothing can be located by selector and clicked
The token is bound to the page URL and the sitekey, so both have to be passed correctly
Tokens expire in about two minutes, so it has to be produced close to the moment of submission
Sites that raise their threshold may still reject a low score, which looks like a failed solve but is not one

The two ways to call it

Token endpoint: you send sitekey, url and action, and a finished token comes back. This is the normal route for v3 and what the example below uses.
Image endpoint: only applies to v2 style challenges. It returns what to click rather than a token, so it is the wrong tool for v3.

Parameters that matter

sitekey: the value of the data-sitekey attribute on the page, required
url: the full page URL where the captcha is rendered, required and must match
action: the action name the page passes to grecaptcha.execute, when the site uses one
proxy: optional. Pass a proxy when the target refuses datacenter addresses
useragent: optional. Send the same one your browser or scraper uses

Where the token goes

The token is written into the hidden field that the page already has, usually named g-recaptcha-response, and then the form is submitted. With a few frameworks you instead hand it to the page callback, for example grecaptcha callback functions or a framework specific resolver. Either way the token is the only thing you need from us.

How billing works here

One accepted token costs one credit, and a submission that the upstream refuses is not charged. Polling for the result is free. A new account starts with 200 free solves, which is roughly 200 tokens, enough to validate an integration end to end before paying anything.

How to integrate

curl -X POST https://fuckcaptcha.top/v1/token/recaptcha3 \\
  -H "Authorization: Basic $NSK_KEY" \\
  -d '{"sitekey":"<sitekey>","url":"<page url>"}'

Pricing

Charged per solve: one credit for each recognition request, and polling for the result is free. Signing up includes 200 free solves.

FAQ

Is reCAPTCHA v3 the same as reCAPTCHA v2?

No. Version 2 shows a checkbox or an image grid and can be solved by recognition. Version 3 shows nothing and returns a score, so it can only be solved by producing a token. They use different endpoints, so picking the wrong one is the most common integration mistake.

What score do I get?

The token is what the page needs. Scores are assigned by Google when the page validates the token and can vary with traffic and reputation, so a token that validates can still be rejected by a site that has raised its own threshold.

How long is a token valid?

Around two minutes. Request it close to the moment of submission instead of caching it, and request a fresh one for each form you fill.

Do I need to pass the action value?

Only when the page calls grecaptcha.execute with an action name. Check the page source first; sending a wrong action can make the token fail validation.

Do you need my cookies or browser session?

No. The token is produced from the sitekey and URL, so no cookies or account access are required.

Read the API reference Download the Chrome extension