Firewall

Block bots your way.

Compose rules from 12 condition types, 6 action types, AND / OR logic — evaluated in under 3 ms per request, at any scale.

Custom Rules

Block by IP, CIDR, ASN, country, user-agent pattern, bot score, or category.

Allow Lists

Whitelist Googlebot, office IPs, monitoring services, always highest priority.

Quick Presets

One-click: Block datacenter IPs, Block AI crawlers, Challenge suspicious.

Industry Templates

Pre-built rule sets for Publishers, E-Commerce, SaaS, and API protection.

Priority System

Rules evaluated by priority. First match wins. Allow rules override blocks.

Scheduled Rules

Activate rules only during specific hours or days.

5 Actions

Block (403), Challenge (PoW), Throttle (rate limit), Log, or Allow.

Hit Tracking

Every rule tracks hit count and last triggered time.

9
Match operators
6
Action types
<3ms
Rule evaluation
Rules per site
How it works

Three minutes from idea to enforcement.

Rules live in your dashboard, version-controlled with a full audit trail. Edits propagate to every site within seconds.

Step 01

Define conditions

Combine IP, CIDR, ASN, country, user-agent regex, bot category, score range, request path, or referer with AND / OR logic.

Step 02

Pick an action

Allow, log, throttle, challenge (CAPTCHA / PoW), redirect, or block. Set TTL per rule for time-boxed enforcement.

Step 03

Watch it work

Every match is logged with full context — score, signals, geo, ASN — so you can audit and tune without guessing.

Worked example

A targeted scraper rule, in plain English.

Block requests that are clearly a competitor scraping product data from AWS, using python-requests, and already flagged as bot — but only on the products API.

RULE · Block AWS python scrapers on /api/v1/products
  • NetworkASN 14618 — Amazon AWSmatch
  • Requestpath matches ^/api/v1/products/.*match
  • Identityuser-agent contains 'python-requests'match
  • Identityscore >= 0.70match
  • ActionCapped at 1.0 — blockedBLOCK
Conditions

12 ways to identify a request.

Mix and match. Every condition is indexed so even complex AND / OR trees evaluate in the same low-millisecond budget.

Network4 signals
IP addressexact

Single IP or comma-separated list. Matches exact.

CIDR rangerange

IPv4 + IPv6 supported. Subnet matching at line speed.

ASNexact

Match by autonomous system number — block entire datacenter or proxy network.

Countryexact

ISO-3166 country code. Geolocation cached, refreshed weekly.

Identity4 signals
User-agentregex

Exact match, contains, or regex. Anchor with ^ and $ for tight matching.

Bot categorytag

ai_crawler · scraper · seo · social · monitoring · search · custom.

Bot scorecompare

Threshold comparison. e.g. score >= 0.7 captures bots, score < 0.3 captures verified humans.

Verified botboolean

rDNS-confirmed identity (Googlebot, Bingbot, Applebot). Separate from claimed UA.

Request4 signals
Path patternregex

Glob or regex match against URL path. Combine with method for tight scoping.

Methodexact

GET, POST, PUT, DELETE, etc. Lock down write endpoints separately.

Refererregex

Match origin domain. Useful for hotlink protection + scraper detection.

Header valueregex

Match any HTTP header by name + value. Detect framework signatures.

Actions

Pick the right response.

A hard block isn't always right. Throttle abusive crawlers, challenge ambiguous traffic, log new rules before flipping them live.

01

Allow

Explicit allow-list. Bypasses all downstream rules. Useful for verified bots + internal tools.

02

Log only

Mark the request but don't intercept. Build confidence in a rule before flipping it to block.

03

Throttle

Reduce request rate per IP / ASN. Configurable RPS, burst, and backoff curve.

04

Challenge

Issue CAPTCHA, proof-of-work, or Cloudflare Turnstile. Human passes silently, bot stalls.

05

Redirect

302 to a honeypot, paywall, or robots.txt explanation. Useful for paid AI-crawler flows.

06

Block

Hard 403. Returns a custom message or template. Logged with full request context.

53% of internet traffic is automated.
How much of yours?

Most site owners have no idea. Find out in under 2 minutes — free.