Skip to main content

Understand your traffic

Every fetch classified: humans, crawlers, scrapers, scanners, AI agents, forgeries. Observe first — nothing is blocked until you say so.

Book a demo →

100,000 events a month. No card. DNS at the edge or a PHP agent on origin.

Botscope actively blocks malicious bots while giving you deep analytics into AI agents like GPTBot, ClaudeBot, and Gemini.

this visit · classifying scanning
You Reading signals from this browser session
Pending
Refining incognito signal — IndexedDB commit timing

This is your visit on this page — not a mock-up.

works with any cms, framework, or custom stack

WordPress WordPress
Laravel Laravel
PHP PHP
React React
Vue.js Vue.js
Next.js Next.js
Django Django
Node.js Node.js
Shopify Shopify
Nuxt Nuxt
TypeScript TypeScript
Drupal Drupal
WordPress WordPress
Laravel Laravel
PHP PHP
React React
Vue.js Vue.js
Next.js Next.js
Django Django
Node.js Node.js
Shopify Shopify
Nuxt Nuxt
TypeScript TypeScript
Drupal Drupal

Check your robots.txt

A plain GET against your domain — no account, no script. Every line of the answer is checkable.

robots.txt · audit ready

Shows which AI crawlers your robots.txt allows, blocks, or never mentions.

01 · The blind spot

Two traffic problems. Both invisible to GA4.

Traditional analytics only show you human traffic. The rest is quietly consuming your resources.

Bad bots — block them

Bots are corrupting your data

Scrapers, headless browsers, and botnets inflate your sessions, poison your SEO signals, and eat your ad budget. GA4 cannot tell you they're there.

  • Fake conversions in your analytics
  • Ad clicks that cost you real money
  • Behavior signals that drag your rankings down
AI agents — track them

AI agents are your next traffic source

GPTBot, ClaudeBot, Gemini — they're crawling your pages right now. When they cite you, you get traffic. But you can't optimize what you can't see.

  • Which pages ChatGPT and Claude reference
  • How often AI-cited pages convert
  • Which AI operators send you real visitors

02 · what you get

This is the screen you open on Monday

Everything above describes how it works. This is what it produces: one week of a mid-size storefront, every request classified, every verdict explained. Numbers below are from our own demo workspace — no customer is named and nothing here is presented as somebody else's result.

app.botscope.io/traffic — demo workspace live
Total requests
0
8.2 / hour
Human traffic
0%
756 requests
Automated
0%
bots · crawlers · AI
AI agents
0
GPTBot, ClaudeBot, Perplexity
Requests per day by class Human Bots AI agents Stopped
26
27
28
29
30
31
1
AI citations by platform
ChatGPT 0
Claude 0
Perplexity 0

Which assistants fetched your pages, and how many readers each one sent back. No DNS migration. And through to conversion, not to a request counter.

What the engine decided, and why
googlebot Search crawler 125 rDNS verified Allow
gptbot AI agent 51 Catalogue match Observe
claudebot AI agent 44 Catalogue match Observe
headless-chrome Automation 38 No JS telemetry Challenge
203.0.113.44 Datacenter IP 26 Honeypot hit Block

Demo workspace data. Your own numbers appear within a day of connecting, in observe mode, before anything is ever blocked.

Book a demo →

03 · under the hood

Four layers of defense

Every request runs the full stack before your application is reached — no redirect, no round trip, nothing for a real visitor to wait on.

L1

Network & Crawlers

  • ·IP reputation & threat feeds
  • ·Verified crawler bypass (Googlebot, GPTBot…)
  • ·Fake-crawler detection via rDNS mismatch
L2

Session Validation

  • ·HMAC-signed shield session cookie
  • ·TLS / JA4 fingerprint continuity check
  • ·Silent re-vouch on IP change (150 ms)
L3

Challenge Barrier

  • ·Invisible Proof-of-Work (browser-only)
  • ·Adaptive difficulty by risk score 0–100
  • ·Custom rule engine — block / challenge / observe
L4

Behavioural Traps

  • ·Invisible honeypot links in HTML
  • ·Stateless pixel tracks static-asset loading
  • ·Token-copier & HTML-only scraper detection

03 · under the hood, continued

What gets stopped?

Everything that shouldn't be there.

HB

Headless browsers

Puppeteer, Playwright, Selenium — detected via JS environment probes.

SP

Scrapers & parsers

HTML-only scrapers, API harvesters, price and content bots.

FC

Fake crawlers

Bots spoofing Googlebot — exposed by rDNS / forward-DNS mismatch.

DC

Datacenter & proxy

AWS, Azure, GCP ranges; residential proxies and Tor exit nodes.

AI

AI training scrapers

Unverified or policy-violating LLM data harvesters.

HP

Honeypot triggers

Any client clicking invisible trap links gets an instant permanent ban.

04 · you already have a WAF

Keep Cloudflare. This answers a different question.

A WAF decides whether a request is dangerous. Botscope decides who the client is and what it is doing with your content — including the AI assistants that are now a traffic source rather than a threat. Most customers run both.

WAF / CDN Botscope
Stops volumetric abuse and L7 floods
This is what a WAF is for, and it is better at it. Botscope blocks at the origin or the edge, per agent — it is not a DDoS layer and does not try to be.
~
Tells you which AI agents are on your site
GPTBot, ClaudeBot, PerplexityBot, Gemini and 30+ more, identified by catalogue and verified by reverse DNS rather than by the user agent they claim.
Shows which AI answers cite your pages
Which pages get fetched for an answer, and how many readers come back from it. A WAF has no concept of this.
A separate decision per agent
Allow the search crawler, observe the AI trainer, challenge the scraper — one policy each, kept in sync with robots.txt. A WAF rule set is written per pattern, not per identity.
~
Bot traffic as analytics, not just as a log
What automated clients read, how often, and what it costs you — in a dashboard rather than in a request log you have to query.
Runs in observe mode before it blocks anything
Every new site starts in observe: it classifies and reports, and enforces nothing until you turn it on. Your first week carries no risk of blocking a customer.
~

If a WAF already covers what you need, that is a fine answer — say so on a call and we will tell you whether this adds anything for you. Book 30 minutes →

05 · ways to connect

Ways to connect

Pick what fits your stack. Switch modes without reinstalling.

JS

Observe JS

Start here — one script tag

Paste one script tag — no server change, no blocking. Streams browser signals and AI-citation tracking; add DNS or the Agent later for enforcement.

No blocking Zero server changes LLM referral tracking
WP

WordPress plugin

A zip, uploaded in wp-admin

The origin agent with the install done for you: no FTP, no php.ini, no shell. It places itself before WordPress boots and reports which position your host allowed.

No server access needed Same agent underneath Observe by default

What it does and does not do →

CF

Cloudflare Worker

Inside your own zone

For a site already behind Cloudflare, whose A record cannot point anywhere else. Runs in front of your origin, enforces there, and reports every decision back.

No DNS change Country and network free Same catalogue and policy

What it does and does not do →

DNS

Protect DNS

Edge proxy on our infra

Point your domain's A record to our edge. All L1–L4 layers run on our proxy — zero load on your server.

Any CMS or language HTML rewrite included Zero server load
PHP

Protect Agent

auto_prepend_file on your server

One line in php.ini or .htaccess. Runs before your app, sends telemetry asynchronously after the response.

PHP 8.1+ Laravel / WP / custom Async telemetry

Want DNS's zero-load edge and a second, HMAC-verified layer of defense in case DNS is ever bypassed? That's not another mode to set up — just also install the Agent on a site already running DNS, and Botscope enables the extra layer automatically. The same is true of the Worker: it shares its secret with the Agent, so a visitor who passed a check at one is not asked again by the other.

06 · pricing

Everything is free. You pay for volume.

Every feature, on every account, up to 100K events a month. Past that, buy capacity in blocks of 100K — change the number any month, or drop back to free.

100K 10M
events / month

No card. Start on the free 100K and add blocks when you outgrow it.

Rate by volume

Per 100K block. The further you go, the less each one costs.

100K – 600K $5.99
600K – 2.1M $3.99
2.1M+ $2.99

Each band is charged at its own rate, so the bill only ever rises with the volume. Increases apply immediately; decreases at your next renewal, with nothing refunded and nothing owed.

Included free, and at every size

Full enforce mode + honeypots Session protection (L2) AI citation & crawler analytics Every known bot & AI agent tracked REST API access Observe mode — nothing blocked until you say so

One billable event = one telemetry record stored (typically one protected HTTP request).

$ botscope --enable

See what's really on your site — free, in under 5 minutes

No credit card, and no expiry. Every feature on every account, free up to 100K events a month. Connect via JS snippet or PHP agent — you pay only if you outgrow it.

Book a demo →