NeuroCaptcha

RecaptchaV3TaskProxyless

Parameters, request and response for reCAPTCHA v3 without a proxy: the RecaptchaV3TaskProxyless task type.

এই বিভাগটি ইংরেজিতে দেখানো হয়েছে।

এই পেজে

reCAPTCHA v3: no widget, the page calls grecaptcha.execute(sitekey, {action}) and the site gets a score. The task is solved from our IP addresses — no proxy needed.

Price0.90 US$ per 1000 tasks (0.0009 US$ per task), only for solved tasks — pricing
Proxynot needed
Queue in getQueueStats18, 19, 20
Also accepted as—

Parameters#

Fields of the task object in createTask:

type
stringআবশ্যক
RecaptchaV3TaskProxyless
websiteURL
stringআবশ্যক
Full address of the page where the reCAPTCHA is shown: http(s), up to 2048 characters. The page, not the address of the reCAPTCHA iframe.
websiteKey
stringআবশ্যক
Site key: the data-sitekey attribute of the widget or the k parameter of the anchor request in DevTools → Network. Up to 256 characters, no spaces.
pageAction
string
The action the page passes to grecaptcha.execute(sitekey, {action: …}), e.g. login. Up to 128 characters.
minScore
number
Minimum score of the token from 0 to 1, usually 0.3, 0.7 or 0.9.
apiDomain
string
Domain the widget is loaded from: www.google.com (default) or www.recaptcha.net.
userAgent
string
User-Agent of the browser that will submit the token, up to 512 characters. It is returned in solution.userAgent.
cookies
string
Cookies of the target site for the solving browser: name=value; name2=value2, up to 8192 characters. Stored encrypted.

Request#

POST /createTask
{
  "clientKey": "YOUR_API_KEY",
  "task": {
    "type": "RecaptchaV3TaskProxyless",
    "websiteURL": "https://example.com/login",
    "websiteKey": "YOUR_SITE_KEY",
    "pageAction": "login",
    "minScore": 0.7
  }
}

Complete example#

Create the task, poll every 2 seconds, print the token:

import os
import time

import requests

API = "https://neurocaptcha.com"
KEY = os.environ["API_KEY"]


def call(method, payload):
    data = requests.post(f"{API}/{method}", json={"clientKey": KEY, **payload}, timeout=30).json()
    if data.get("errorId"):
        raise RuntimeError(f'{data["errorCode"]}: {data.get("errorDescription", "")}')
    return data


task = {
    "type": "RecaptchaV3TaskProxyless",
    "websiteURL": "https://example.com/login",
    "websiteKey": "YOUR_SITE_KEY",
    "pageAction": "login",
    "minScore": 0.7,
}

# 1. create the task: its price is held, not charged
task_id = call("createTask", {"task": task})["taskId"]

# 2. poll every 2 s; a failed task raises with its error code and costs nothing
deadline = time.time() + 300
while time.time() < deadline:
    time.sleep(2)
    result = call("getTaskResult", {"taskId": task_id})
    if result["status"] == "ready":
        print(result["solution"]["gRecaptchaResponse"])
        break
else:
    raise TimeoutError("the task was not solved in time")

Response#

getTaskResult once the task is solved:

getTaskResult
{
  "errorId": 0,
  "status": "ready",
  "solution": {
    "gRecaptchaResponse": "03AFcWeA6Z3vq0VhGm1oS8r0bAxKk4…",
    "userAgent": ""
  },
  "cost": "0.000900",
  "ip": "198.51.100.7",
  "createTime": 1793370000,
  "endTime": 1793370007,
  "solveCount": 1
}

Tips#

  • Find the action in the page code: grecaptcha.execute('<sitekey>', {action: 'login'}). A wrong action gives a token the site rejects.
  • Ask for the score the site really needs: a higher minScore can take longer.
  • The token comes in solution.gRecaptchaResponse; the score itself is not returned — the site gets it from Google when it verifies the token.
  • If the site binds tokens to the IP address, switch to a type with your proxy.