From be329447abdc05cf3e822bfc1d0d1935f0e3c57a Mon Sep 17 00:00:00 2001
From: Daniel Nitsikopoulos <daniel@dnitza.com>
Date: Sun, 16 Jun 2024 09:20:32 +1000
Subject: [PATCH] Block AI bots

---
 Gemfile      | 1 +
 Gemfile.lock | 3 +++
 config.ru    | 7 +++++++
 3 files changed, 11 insertions(+)

diff --git a/Gemfile b/Gemfile
index de97a9f..1abe236 100644
--- a/Gemfile
+++ b/Gemfile
@@ -38,6 +38,7 @@ gem "matrix"
 gem "rack-session"
 gem "rack-rewrite"
 gem "rack-contrib"
+gem "rack-attack", "~> 6.7"
 
 gem "ruby-readability", require: "readability"
 gem "down"
diff --git a/Gemfile.lock b/Gemfile.lock
index 27cd766..9c5068c 100644
--- a/Gemfile.lock
+++ b/Gemfile.lock
@@ -365,6 +365,8 @@ GEM
     que (2.3.0)
     racc (1.8.0)
     rack (2.2.9)
+    rack-attack (6.7.0)
+      rack (>= 1.0, < 4)
     rack-contrib (2.5.0)
       rack (< 4)
     rack-rewrite (1.5.1)
@@ -579,6 +581,7 @@ DEPENDENCIES
   pg
   puma
   que
+  rack-attack (~> 6.7)
   rack-contrib
   rack-rewrite
   rack-session
diff --git a/config.ru b/config.ru
index 872d471..c71d8d2 100644
--- a/config.ru
+++ b/config.ru
@@ -23,6 +23,13 @@ use Rack::Rewrite do
   }
 end
 
+require "rack/attack"
+use Rack::Attack
+
+Rack::Attack.blocklist("block AI") do |req|
+  req.user_agent.match /AdsBot-Google|Amazonbot|anthropic-ai|Applebot|Applebot-Extended|AwarioRssBot|AwarioSmartBot|Bytespider|CCBot|ChatGPT-User|ClaudeBot|Claude-Web|cohere-ai|DataForSeoBot|Diffbot|FacebookBot|FriendlyCrawler|Google-Extended|GoogleOther|GPTBot|img2dataset|ImagesiftBot|magpie-crawler|Meltwater|omgili|omgilibot|peer39_crawler|peer39_crawler\/1.0|PerplexityBot|PiplBot|scoop.it|Seekr|YouBot/
+end
+
 require "adamantium/middleware/header_fix"
 use Adamantium::Middleware::HeaderFix do |headers, env|
   unless headers["Content-Type"]&.downcase&.include?("xml") || headers["Content-Type"]&.downcase&.include?("json")