Get a page past a bot check, captcha or fingerprint test
One company's docs show a full fit.
What counts as done
The target page loads and the agent can act on it, and the solve happens without a human in one session.
Measured, for our own tests: the check does not return on the next click. No docs we read state these figures, so they don't count against a listing here.
How agents ask for it: the site shows a captcha, get past it
, it keeps blocking me, use a stealth browser
, search Google without hitting the captcha
Does the job 1
The docs show it meets the test above, and nothing we read says otherwise.
Does the job, with a limit 11
It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.
| Listing | Why |
|---|---|
| Airtop C Switchboard Visual Technologies, Inc. (dba Airtop) | solveCaptcha on session creation turns on a built-in solver with onCaptchaEvent callbacks, and a residential proxy is a session option. No tier is named for the solver and no first-hand test exists; the docs point to CapSolver-style extensions for more flexibility. source |
| Anchor Browser B Anchor Forge Inc. | captcha_solver needs an active proxy and is on Starter and above; extra_stealth is Growth only and switches itself off without a proxy. acc-35's free-tier test found the captcha solver failed. source |
| Browser Use A Browser Use, Inc. | Stealth, residential proxies and captcha solving are on by default on every plan, with the docs saying access to every site is not guaranteed and to wait about 10 seconds for the solver. acc-29 hit Google captchas after the 0.7.7 library change; acc-27 and acc-68 were closed not planned. source |
| Browserbase BB Browserbase, Inc. | Captcha solving is No on Free and Auto on Developer and above, taking up to 30 seconds, with Verified fingerprints on Scale. acc-35 found standard mode detected and captcha solving paid; acc-36 saw a press-and-hold check pass the first page and return on the search click. source |
| Browserless BB browserless.io, Inc. | /stealth routes, solveCaptchas=true on the connection URL and the BQL solve mutation cover Cloudflare, reCAPTCHA, DataDome and Akamai at 10 units a solve, with the MCP solve tool marked experimental. acc-66 was still blocked by signature on stealth; acc-35 hit 429 on three parallel sessions. source |
| Hyperbrowser B S2 Labs Inc. | useStealth and ultra stealth modes, solveCaptchas and proxies are session parameters, but captcha solving and proxies need a paid plan, and no first-hand account tests them. source |
| Kernel B Kernel Technologies, Inc. | Anti-detection is on by default and stealth: true adds a static-IP proxy and an automatic captcha solver for reCAPTCHA and Cloudflare; Web Bot Auth is Startup and Enterprise on request. acc-36 saw stealth pass Skyscanner's first page but not the search click. source |
| Pomerado B Pomerado AI, Inc. | The hand-off guide says Pomerado never hands off for a captcha, implying it clears them itself, but no stealth, proxy or solver is documented and no account tests it. source |
| Skyvern C Ikonomos Inc. | A Stealth Chromium browser type and a US residential proxy by default, with a consistent IP option for logins; the Captcha Solver extension is Enterprise only. source |
| Steel BB Nen Labs, Inc. | One session flag turns on detection, solving and verification of captchas at $3 per 1,000 on Launch, with a $10 paid balance required, and the Stealth Browser is Enterprise only. No first-hand pass or fail account exists. source |
| TinyFish B Tiny Fish, Inc. | browser_profile: stealth and proxies are parameters and the anti-bot guide walks through confirming a block, but no captcha solver is described and no first-hand account tests it. source |
Doesn't do this job 10
Listings in the category that do something else, with the reason.
| Listing | Why |
|---|---|
| agent-browser B Vercel Inc. | No stealth or captcha solving in core; the stealth request was closed as not planned (acc-56). The plugins page leaves both to third-party launch.mutate and captcha.solve plugins and names none that ship. source |
| BxPlaywright C Ortus Solutions, Corp. | No stealth, proxy or captcha feature in the module's pages. source |
| Chrome DevTools MCP BB Google LLC | No stealth, proxy or captcha feature in the tool reference or configuration; it drives a plain Chrome. source |
| CurlShot D MEDIAPRONET SRL | No stealth, proxy or captcha parameter in the OpenAPI document or agent guide. source |
| Functionize MCP server D Functionize, Inc. | No stealth, proxy or captcha feature in the capabilities or limitations references. source |
| Kane CLI C TestMu AI (formerly LambdaTest) | No stealth, proxy or captcha feature in the user guide. source |
| Lightpanda C Selecy SAS | acc-65's bot-detection request was closed as not planned; Web Bot Auth signs requests as a declared bot rather than hiding one. source |
| Playwright MCP B Microsoft Corporation | No stealth, proxy or captcha option among the server flags; it launches a plain Playwright browser. source |
| Puppeteer BB Google LLC | No stealth, proxy or captcha feature in Puppeteer itself. source |
| Selenium BB Software Freedom Conservancy | No stealth, proxy or captcha feature in the project; the docs discourage automating Gmail and Facebook logins. source |
What stops agents finishing
- stealth requests closed as not planned on open-source tools (acc-29, acc-56, acc-65)
- a documented captcha solver that fails in a test (acc-35)
- a check that returns on the next click after stealth passed the first page (acc-36)
- captcha solving on paid tiers only (acc-35)
Around it
Before: web-scrapers.protected-page-fetch, browser.read-dynamic-page. After: browser.read-dynamic-page, browser.multi-step-form.
Where this comes from
Weight 8 of 100 in Browser automation: 9 of 61 accounts (acc-22, acc-27, acc-29, acc-35, acc-36, acc-56, acc-65, acc-66, acc-68); mostly GitHub issues on the open-source tools asking for stealth
The same as data: /api/v1/use-cases/browser.pass-bot-check.json, or the MCP tool get_use_case with id browser.pass-bot-check. Other jobs in this category: /api/v1/niches?category=browser.
For companies
Do agents find, use and choose your tools?
An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.