REAL ANTI-DETECT BROWSER — NOT A STEALTH PLUGIN

The identity layer
for AI agents.

One API call. A CDP-attached anti-detect browser in 2 seconds. Burner or persistent — same SDK, same endpoint.

Start free → 100 credits Read the docs
No credit card. Provision your first burner in 90 seconds.
# Provision a 5-minute burner identity. Returns a CDP URL you
# drop straight into Playwright / Puppeteer / your own CDP client.
curl -X POST https://agents.neout.com/v1/agents \
  -H "X-API-Key: $AGENTS_KEY" \
  -H "Content-Type: application/json" \
  -d '{"ttl":"5m","country":"us"}'

# → 201 Created
{
  "id":           "agt_01HXYZ...",
  "cdp_url":      "wss://app.neout.com/api/v1/cdp/<profile>/ws/page/...",
  "expires_at":   "2026-06-05T09:34:00Z",
  "ttl_seconds":  300,
  "trust_score":  760,
  "cost_credits": 1
}
# pip install neout-agents
from neout_agents import Client

c = Client(api_key="agt_live_...")

# Run a browser-driven AI task — returns when done.
# Real anti-detect Chromium + Claude Opus + 5-second live noVNC stream.
job = c.run_task(
    task="Go to news.ycombinator.com and return the top story title.",
    llm="claude-opus-4-7",
    llm_api_key="sk-ant-...",
)
print(job.final_result)
print("watch live:", job.live_url)
# → "Show HN: Self-host…"  + live stream URL while it ran
// npm i @neout/agents
import { Client } from "@neout/agents";

const c = new Client({ apiKey: "agt_live_..." });

// Same one-shot API. Returns when the agent finishes.
const job = await c.runTask({
  task: "Go to news.ycombinator.com and return the top story title.",
  llm: "claude-opus-4-7",
  llmApiKey: "sk-ant-...",
});
console.log(job.finalResult);
console.log("watch live:", job.liveUrl);
// Returns the actual top-story title as text.
// Claude Desktop config (claude_desktop_config.json)
// Once added, ask Claude: "go to news.ycombinator.com and tell me the top story"
{
  "mcpServers": {
    "neout-agents": {
      "command": "npx",
      "args": ["-y", "@neout/mcp-agents"],
      "env": {
        "NEOUT_API_KEY": "agt_live_...",
        "NEOUT_LLM_API_KEY": "sk-ant-..."
      }
    }
  }
}

// Claude Desktop now has 9 tools: run_task, get_job, cancel_job, respond_to_job,
// credits, list_webhooks, create_webhook, delete_webhook, list_identities.

Why this is different.

Every other "AI agent browser cloud" — Hyperbrowser, Browserbase, Browserless — runs stealth plugins on stock browsers. We don't. The browser itself is the moat.

01

A real anti-detect browser

50+ patches deep in the browser engine: navigator.webdriver IDL-removed, WebGL vendor/renderer Apple-shape, audio entropy, canvas noise, NaN canonicalization, shader precision. CreepJS reports our cloud as 19% headless. Vanilla Playwright reports 94%. That gap is why your agent doesn't get walled.

02

One API, two modes

Burner for one-shot tasks, persistent for stateful workflows. Same endpoint, same SDK, same CDP URL pattern. Your agent picks per-call. No "session vs profile" mental tax like the alternatives — just POST /v1/agents with the TTL you want.

03

Trust score on every response

0-1000 score derived from fingerprint coherence, isolation audit, behavioral history, detection-harness verdicts. Tells you which identities are healthy and which are about to get flagged — before the first detection event. No other vendor does this.

Two modes. One API.

Customer decides per-call. Same auth, same SDK, same CDP endpoint, same dashboard, same bill.

BURNER

Ephemeral.

Fresh identity in 2 seconds, self-destructs in 5–60 minutes. Pay per call. No correlation history. Perfect for one-shot AI agent tasks: fill a form, scrape a page, run a check.

POST /v1/agents
{ "ttl": "5m" }

→ 1 credit ($0.10)
  • Fresh identity per call, never reused
  • 5min – 1hr TTL, customer picks
  • No state, no vault, no correlation
  • Auto-destructs on TTL
PERSISTENT

Durable.

Long-lived identity with state: cookies, localStorage, login session, browsing history. For agents that need to remember things across runs. Trust score grows as identity ages — like a real user.

POST /v1/agents
{ "persona": "us-shopper-30s",
  "warmup": "amazon" }

→ $5/identity/month
  • Cookies + localStorage persist across calls
  • Cookie Robot warming, optionally on a schedule
  • Trust score grows with age
  • Re-attach via POST /v1/agents/:id/sessions

Everything you can do.

One product, six entry points. Start where you are.

Dashboard

Mint API keys, watch jobs land, register webhooks, download CSV/NDJSON. Sign in once.

Quickstart

Pick a language, pick a library, copy three blocks. First burner in 90 seconds.

{ }

API Reference

Every endpoint, every field, every error code — with copy-ready curl, Python, and TypeScript snippets.

</>

Examples

Real customer code: scraping, form filling, monitoring, account ops, Amazon SERP, Claude task runner.

MCP Server

Drop into Claude Desktop, Cursor, Continue. 8 tools. Zero code to drive browsers from your AI assistant.

UNIQUE

Live Benchmarks

Open-source detection harness — same five sites, same hardware, run it yourself. No vendor publishes this.

Real numbers, same minute.

Our detection harness runs every Monday. Below: anti-detect ON (neout/agents) vs OFF (vanilla Playwright on the same hardware). Run the harness yourself — it's open-source. Live leaderboard →

Detection site neout/agents Vanilla Playwright
same hardware, no anti-detect
CreepJS
% like headless (lower = better)
19% 94%
fingerprint.com BotD ✓ pass ✗ bot:true
antoinevastel "are you headless" NOT_HEADLESS HEADLESS_DETECTED
amazon.com search cards visible "Sorry, something went wrong"
google.com search results render /sorry/ interstitial

Public head-to-head against Hyperbrowser / Browserbase / Browserless rolls out as the harness picks them up — each one runs against the same 5 detection sites on the same hardware. Source: services/detection-harness on GitHub. Reproduce with one command.

Proof
Reproducible Updated weekly
19%
CreepJS reports our cloud agents as 19% headless. Vanilla Playwright reports 94%. That gap is why your scraper doesn't get walled.
50+
Patches deep in the browser engine — navigator.webdriver, canvas noise, audio entropy, WebGL Apple-shape, shader precision. Not JS shims.
2s
Median time from POST /v1/agents to a CDP-ready browser. Provision, attach proxy, warm-up, return. Every region.

Three minutes to your first agent.

Pick a language, pick a library, copy three blocks. That's it.

1

Install

One pip command. The SDK has zero runtime dependencies — pure stdlib.

$ pip install neout-agents
2

Run a task

Five lines. Get back the agent's result + a live noVNC stream URL.

from neout_agents import Client

c = Client(api_key="agt_live_...")
job = c.run_task(
    task="Top HN story title.",
    llm="claude-opus-4-7",
    llm_api_key="sk-ant-...",
)
print(job.final_result)
3

Drive it

Your existing Playwright code works unchanged — just swap chromium.launch() for chromium.connect_over_cdp(agent.cdp_url).

with sync_playwright() as p:
    browser = p.chromium.connect_over_cdp(agent.cdp_url)
    page = browser.new_page()
    page.goto("https://news.ycombinator.com/")
    print(page.title())
    page.screenshot(path="hn.png")

Pricing.

Credit packs that never expire. Failed provisions auto-refund. No subscription needed for burner.

Free
$0 · 100 credits
For your first agent. No card required.
Starter
$20 · 200 credits
Pay-as-you-go. Country-routed proxies.
See full pricing → Pro pack $500 → 7,500 credits · Enterprise on request

Honest answers.

The questions everyone asks but no anti-detect vendor wants to answer publicly.

Is this just stealth-plugin with extra steps?

No. The browser running underneath neout/agents is purpose-built for anti-detect: 50+ patches deep in the engine, not JS injected at page load. Stealth plugins are detectable by anyone running their own anti-cheat (they're JS sitting in the global scope). Our changes happen at the layer where fingerprinting actually originates — well before any page script runs. See the detection benchmark.

Why per-call credits instead of a monthly subscription?

Because most AI agents have spiky usage — 5 minutes of demand, then nothing for 3 days, then 200 calls in an hour. A flat-rate $50/month subscription either over-charges the slow developer or under-charges the hot startup. Credits map to actual usage cleanly. Persistent identities (P2) are subscription-billed because they cost us money every day they exist, not just when you call them.

What's the difference vs Hyperbrowser / Browserbase?

Functionally similar API surface. The underlying browser is different: ours is a purpose-built anti-detect browser; theirs are stealth plugins on stock browsers. The difference shows on every detection benchmark — Cloudflare, Stripe, Amazon, Booking, Akamai-protected sites all behave differently. We also publish public benchmarks; they don't.

Will my Playwright code work?

Yes. Replace chromium.launch() with chromium.connect_over_cdp(agent.cdp_url), drop the headless/proxy/UA flags (we handle those), and your existing scripts run unchanged. Puppeteer + Selenium + raw CDP also work — anything that speaks the Chrome DevTools Protocol.

What countries can I route through?

Default routes through neout's US-pooled proxies (no charge). Specify country in the request (e.g. "uk", "de", "fr") for country-targeted routing — costs +1 credit per call. Country-specific proxies cover 190+ regions via integrated providers.

What happens if the target site detects my agent?

The agent run continues — we don't crash on a 403. You'll see the bot-wall response in your code and can decide how to handle it (retry with a new burner, use a country-routed proxy, escalate). The trust score on the response is our best estimate of detection risk. P3 adds detection-event webhooks so we can tell you proactively when a persistent identity gets flagged.

How is this different from your existing neout product?

The core neout product is a desktop anti-detect browser for managing many accounts manually. neout/agents is the same anti-detect tech exposed as a per-call API for AI agents — different customer (developers vs operators), different surface (REST + CDP vs desktop launcher), same browser underneath.

Provision your first agent in 90 seconds.

100 credits free. No card required. POST /v1/agents returns a CDP URL you can drop into Playwright today.

Start free → 100 credits