Solve reCAPTCHA v3 (and v3 Enterprise) challenges. Returns a token you verify server-side via siteverify with your secret.

reCAPTCHA v3

reCAPTCHA v3 runs invisibly — there is no checkbox or puzzle. The page calls grecaptcha.execute(site_key, { action }), Google scores the session, and hands back a token. The solver drives that flow in a real browser behind your proxy and returns the resulting token. You then verify it server-side with your secret key; the score and action checks are yours to make.

Request

Field	Type	Required	Notes
`type`	string	Yes	`recaptchav3` (or alias `recaptcha-v3`).
`site_key`	string	Yes	The reCAPTCHA site key (the `render`/`data-sitekey` value, usually starts `6L`).
`url`	string	Yes	The page where reCAPTCHA v3 runs. The token is bound to this URL's domain.
`action`	string	No	The action name to execute. Defaults to `verify` if omitted — match what your backend asserts.
`enterprise`	bool	No	`true` for reCAPTCHA Enterprise (loads `enterprise.js`). Defaults to `false` (standard v3).
`proxy`	string	Yes	A proxy is required. `http://`, `https://`, `socks4://`, `socks5://` schemes only.
`user_agent`	string	No	Optional User-Agent override for the solving browser.

Unlike Turnstile, proxy is required. reCAPTCHA scores the session partly on the egress IP, so the solver routes the whole flow through your proxy. Use a clean IP — a poor reputation egress yields a low score on siteverify.

Response

{
  "success": true,
  "type": "recaptchav3",
  "token": "03AFcWeA...",
  "cost": 0.0015
}

Field	Type	Notes
`success`	bool	Always `true` on 200.
`type`	string	Echoes the request type (`recaptchav3`).
`token`	string	The reCAPTCHA response token. Submit it where your app expects `g-recaptcha-response`.
`cost`	number	USD deducted from your balance for this call.

Forward the token to your backend and verify it against Google's siteverify endpoint with your secret key. The token is bound to the domain of the url you submitted — the hostname field in the siteverify response will match it. The returned score and action are for you to evaluate against your own thresholds; the API does not gate on them.

Verify server-side

A token on its own proves nothing. Always call siteverify from your server with your secret, then check success, hostname, action, and score before trusting the request.

How it works (under the hood)

The solver opens the target url in a real browser context, routed through your proxy.
It loads reCAPTCHA's api.js (or enterprise.js when enterprise is true) for your site_key.
It calls grecaptcha.execute(site_key, { action }) — using action if you sent one, otherwise the default verify.
Google issues a scored token, which the solver returns.

You don't drive any of this — you just need site_key, url, and proxy.

Performance

Metric	Typical value
Solve time	3-6 seconds
Success rate	90%+
`score` returned	0.7-0.9 with a clean proxy

Solve time is dominated by browser startup and the proxy latency on each leg. The score you get back from siteverify depends mostly on the egress IP reputation of your proxy — residential/mobile IPs score highest, datacenter IPs lowest.

Examples

import requests

proxy = "http://user:[email protected]:8000"

r = requests.post(
    "https://api.nslsolver.com/solve",
    headers={"X-API-Key": "YOUR_KEY"},
    json={
        "type": "recaptchav3",
        "site_key": "6LcAAAAA...",
        "url": "https://www.target.com/login",
        "action": "login",
        "proxy": proxy,
    },
    timeout=120,
)
solved = r.json()
token = solved["token"]

# Verify server-side with YOUR secret before trusting it
verify = requests.post(
    "https://www.google.com/recaptcha/api/siteverify",
    data={"secret": "YOUR_SECRET", "response": token},
).json()
# verify["success"], verify["hostname"], verify["action"], verify["score"]

const proxy = "http://user:[email protected]:8000";

const r = await fetch("https://api.nslsolver.com/solve", {
  method: "POST",
  headers: {
    "Content-Type": "application/json",
    "X-API-Key": process.env.NSL_API_KEY,
  },
  body: JSON.stringify({
    type: "recaptchav3",
    site_key: "6LcAAAAA...",
    url: "https://www.target.com/login",
    action: "login",
    proxy: proxy,
  }),
  signal: AbortSignal.timeout(120_000),
});

const { token } = await r.json();
// Submit `token` as g-recaptcha-response, then siteverify on your backend.

type RecaptchaReq struct {
    Type     string `json:"type"`
    SiteKey  string `json:"site_key"`
    URL      string `json:"url"`
    Action   string `json:"action,omitempty"`
    Proxy    string `json:"proxy"`
}

payload, _ := json.Marshal(RecaptchaReq{
    Type:    "recaptchav3",
    SiteKey: "6LcAAAAA...",
    URL:     "https://www.target.com/login",
    Action:  "login",
    Proxy:   "http://user:[email protected]:8000",
})

req, _ := http.NewRequest("POST", "https://api.nslsolver.com/solve", bytes.NewReader(payload))
req.Header.Set("Content-Type", "application/json")
req.Header.Set("X-API-Key", os.Getenv("NSL_API_KEY"))

client := &http.Client{Timeout: 120 * time.Second}

curl -X POST https://api.nslsolver.com/solve \
  -H "Content-Type: application/json" \
  -H "X-API-Key: $NSL_API_KEY" \
  -d '{
    "type": "recaptchav3",
    "site_key": "6LcAAAAA...",
    "url": "https://www.target.com/login",
    "action": "login",
    "proxy": "http://user:[email protected]:8000"
  }'

For reCAPTCHA Enterprise, add "enterprise": true to any of the requests above.

Troubleshooting

Symptom	Likely cause	Fix
`site_key is required`	Missing site key	Send the `render`/`data-sitekey` value (usually starts `6L`).
`url is required`	Missing URL	Send the page URL where reCAPTCHA v3 runs.
`proxy is required`	Missing proxy	reCAPTCHA v3 scores partly on the egress IP — a proxy is required.
`proxy format must be protocol://[user:pass@]host:port`	Bad proxy URL	Use `http://`, `https://`, `socks4://`, or `socks5://`.
Low `score` on `siteverify`	Poor proxy IP reputation, or action mismatch	Use cleaner (residential) proxies; ensure `action` matches your check.
`action` mismatch on `siteverify`	Sent action ≠ the one your backend asserts	Send the same `action` your server expects (default is `verify`).
`hostname` mismatch on `siteverify`	Solved a different domain than you verify	Use a `url` on the same domain you submit the token against.
Enterprise token rejected	Site uses `enterprise.js` but solved standard	Set `"enterprise": true`.

reCAPTCHA v3

reCAPTCHA v3

Request

Response

How it works (under the hood)

Performance

Examples

Troubleshooting

On this page