3-day free trial · no card required · Keitaro migration in 24 hours
Try
Bot protection

Bot protection with granular controls.

A 3-layer antibot: a rule-based filter with custom IP/UA white and black lists, a fingerprint engine of ~40 signals across IP, behavior, fingerprint and technical, and an ensemble scorer that produces a final probability. You choose the filter level and where detected bots go — not a black box, full control.

Start free Compare with Keitaro

How antibot works on TDS

1

Rule filter and lists

A fast rule-based layer runs first: custom IP/UA white and black lists, a white-referer allowlist, empty-UA and local-IP checks, ISP and hostname signatures. Filter levels 0–3 control strictness, from common checks to strict VPN/server/country signatures.

2

IP and network checks

Each click is scored on its network: datacenter orgs (AWS, Google Cloud, Azure, Linode), datacenter hostnames, TOR exit nodes, public/web/residential proxies, VPN with timezone-mismatch and an IP-reputation list.

3

Fingerprint and behavior

~40 signals across fingerprint and behavior: WebGL/canvas, headless GPU markers (swiftshader/llvmpipe), fonts, audio, instant-post timing and mouse-trail jitter. navigator.webdriver and automation tools (Selenium, Puppeteer, Playwright) are hard signals — near-certain on their own.

4

Ensemble verdict and action

An ensemble scorer combines rule, logistic, behavioral and render-challenge sub-models into one probability: ≥ 0.5 shows a CAPTCHA, ≥ 0.7 is treated as a bot and routed to your safe page — real users continue to the offer.

Why it matters

  • Configurable filter levels (0–3) plus custom IP/UA white and black lists
  • ~40 signals across IP, behavior, fingerprint and technical categories
  • Hard-detects webdriver and automation (Selenium, Puppeteer, Playwright)
  • Bots routed to a safe page — real users reach the offer
~40 signals
0–3 Filter level

Related modules

Traffic filtering → Real-time analytics →

FAQ

How is this different from a black-box antibot?
Every layer is transparent and configurable: the filter level (0–3), the white/black lists and where detected bots are sent. You control the strictness instead of trusting a hidden score.
How are bots actually decided?
Signals aggregate into a probability. At ≥ 0.5 a visitor gets a CAPTCHA challenge; at ≥ 0.7 they are treated as a bot and routed to your safe page; below that they pass through. One weak signal alone (a VPN, Brave, an unusual config) is never enough — a verdict needs several anomalies to coincide.
Do I need my own server?
No. The antibot runs in the cloud on our infrastructure — no VPS, no add-ons.

Open an account.
Decide in three days.

Full access to the Standard plan. No card required. Setup help on Telegram in 15 minutes.

Start free trial Message us on Telegram
Start free trial