breach-mitigation-rails 0.0.1

data/.gitignore

@@ -0,0 +1,17 @@
+*.gem
+*.rbc
+.bundle
+.config
+.yardoc
+Gemfile.lock
+InstalledFiles
+_yardoc
+coverage
+doc/
+lib/bundler/man
+pkg
+rdoc
+spec/reports
+test/tmp
+test/version_tmp
+tmp

data/Gemfile

@@ -0,0 +1,4 @@
+source 'https://rubygems.org'
+
+# Specify your gem's dependencies in breach-mitigation-rails.gemspec
+gemspec

data/LICENSE.txt

@@ -0,0 +1,22 @@
+Copyright (c) 2013 BBA, Inc.
+
+MIT License
+
+Permission is hereby granted, free of charge, to any person obtaining
+a copy of this software and associated documentation files (the
+"Software"), to deal in the Software without restriction, including
+without limitation the rights to use, copy, modify, merge, publish,
+distribute, sublicense, and/or sell copies of the Software, and to
+permit persons to whom the Software is furnished to do so, subject to
+the following conditions:
+
+The above copyright notice and this permission notice shall be
+included in all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE
+LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

data/README.md

@@ -0,0 +1,58 @@
+# breach-mitigation-rails
+
+Makes Rails applications less susceptible to the BREACH / CRIME
+attacks. See [breachattack.com](http://breachattack.com/) for details.
+
+## How it works
+
+This gem implements two of the suggestion mitigation strategies from
+the paper:
+
+*Masking Secrets*: The Rails CSRF token is 'masked' by encrypting it
+with a 32-byte one-time pad, and the pad and encrypted token are
+returned to the browser, instead of the "real" CSRF token. This only
+protects the CSRF token from an attacker; it does not protect other
+data on your pages (see the paper for details on this).
+
+*Length Hiding*: The BreachMitigation::LengthHiding middleware
+appends an HTML comment up to 2k in length to the end of all HTML
+documents served by your app. As noted in the paper, this does not
+prevent plaintext recovery, but it can slow the attack and it's
+relatively inexpensive to implement. Unlike the CSRF token masking,
+length hiding protects the entire page body from recovery.
+
+## Warning!
+
+BREACH and CRIME are **complicated and wide-ranging attacks**, and this
+gem offers only partial protection for Rails applications. If you're
+concerned about the security of your web app, you should review the
+BREACH paper and look for other, application-specific things you can
+do to prevent or mitigate this class of attacks.
+
+## Installation
+
+Add this line to your Rails Gemfile:
+
+    gem 'breach-mitigation-rails'
+
+And then execute:
+
+    $ bundle
+
+TODO And then?
+
+## Gotchas
+
+length padding can screw up etags
+length padding increases your bandwidth, could increase page load
+times
+if you have overridden verified_request? in your app, need to make sure you call +super+
+
+## Contributing
+
+Pull requests are welcome, either to enhance the existing mitigation
+strategies or to add new ways to mitigate against the attack.
+
+## License
+
+MIT - see LICENSE.txt

data/Rakefile

@@ -0,0 +1 @@
+require "bundler/gem_tasks"

data/breach-mitigation-rails.gemspec

@@ -0,0 +1,23 @@
+# coding: utf-8
+lib = File.expand_path('../lib', __FILE__)
+$LOAD_PATH.unshift(lib) unless $LOAD_PATH.include?(lib)
+require 'breach_mitigation/version'
+
+Gem::Specification.new do |spec|
+  spec.name          = "breach-mitigation-rails"
+  spec.version       = BreachMitigation::VERSION
+  spec.authors       = ["Bradley Buda"]
+  spec.email         = ["bradleybuda@gmail.com"]
+  spec.description   = %q{Mitigates the BREACH and CRIME attacks on TLS in Rails applications}
+  spec.summary       = %q{Uses length-hiding and CSRF token masking to make it more difficult for an attacker to recover plaintext from HTTP responses. See README.md for details.}
+  spec.homepage      = "https://github.com/meldium/breach-mitigation-rails"
+  spec.license       = "MIT"
+
+  spec.files         = `git ls-files`.split($/)
+  spec.executables   = spec.files.grep(%r{^bin/}) { |f| File.basename(f) }
+  spec.test_files    = spec.files.grep(%r{^(test|spec|features)/})
+  spec.require_paths = ["lib"]
+
+  spec.add_development_dependency "bundler", "~> 1.3"
+  spec.add_development_dependency "rake"
+end

data/lib/breach-mitigation-rails.rb

@@ -0,0 +1 @@
+require 'breach_mitigation/railtie' if defined?(Rails)

data/lib/breach_mitigation/length_hiding.rb

@@ -0,0 +1,50 @@
+module BreachMitigation
+  class LengthHiding
+    def initialize(app)
+      @app = app
+    end
+
+    def call(env)
+      status, headers, body = @app.call(env)
+
+      # Only pad HTML documents
+      if headers['Content-Type'] =~ /text\/html/
+        # Copy the existing response to a new object
+        response = Rack::Response.new(body, status, headers)
+
+        # Append to that response
+        response.write random_html_comment
+
+        body.close if body.respond_to? :close
+        response.finish
+      else
+        [status, headers, body]
+      end
+    end
+
+    private
+
+    # Append a comment from 0 to MAX_LENGTH bytes in size to the
+    # response body. See section 3.1 of "BREACH: Reviving the CRIME
+    # attack". This should make BREACH attacks take longer, but does
+    # not fully protect against them. The longer MAX_LENGTH is, the
+    # more effective the mitigation is, however longer lengths mean
+    # more time spent in this middleware and more data on the wire.
+
+    MAX_LENGTH = 2048
+    ALPHABET = ('a'..'z').to_a
+
+    def random_html_comment
+      # The length of the padding should be strongly random, but the
+      # data itself doesn't need to be strongly random; it just needs
+      # to be resistant to compression
+      length = SecureRandom.random_number(1024)
+
+      # TODO make this faster
+      junk = ''
+      length.times { junk << ALPHABET.sample }
+
+      "\n<!-- This is a random-length HTML comment: #{junk} -->"
+    end
+  end
+end

data/lib/breach_mitigation/masking_secrets.rb

@@ -0,0 +1,47 @@
+module BreachMitigation
+  class MaskingSecrets
+    class << self
+      AUTHENTICITY_TOKEN_LENGTH = 32
+
+      # Sets the token value for the current session and returns it in
+      # a masked form that's safe to send to the client. See section
+      # 3.4 of "BREACH: Reviving the CRIME attack".
+      def masked_authenticity_token(session)
+        one_time_pad = SecureRandom.random_bytes(AUTHENTICITY_TOKEN_LENGTH)
+        encrypted_csrf_token = xor_byte_strings(one_time_pad, real_csrf_token(session))
+        masked_token = one_time_pad + encrypted_csrf_token
+        Base64.strict_encode64(masked_token)
+      end
+
+      # Checks the client's masked token to see if it matches the
+      # session token. Essentially the inverse of
+      # +masked_authenticity_token+.
+      def valid_authenticity_token?(session, encoded_masked_token)
+        return false if encoded_masked_token.nil? || encoded_masked_token.empty?
+
+        masked_token = Base64.strict_decode64(encoded_masked_token)
+
+        return false if masked_token.length != AUTHENTICITY_TOKEN_LENGTH * 2
+
+        # Split the token into the one-time pad and the encrypted
+        # value and decrypt it
+        one_time_pad = masked_token[0...AUTHENTICITY_TOKEN_LENGTH]
+        encrypted_csrf_token = masked_token[AUTHENTICITY_TOKEN_LENGTH..-1]
+        csrf_token = xor_byte_strings(one_time_pad, encrypted_csrf_token)
+
+        csrf_token == real_csrf_token(session)
+      end
+
+      private
+
+      def real_csrf_token(session)
+        session[:_csrf_token] ||= SecureRandom.base64(AUTHENTICITY_TOKEN_LENGTH)
+        Base64.strict_decode64(session[:_csrf_token])
+      end
+
+      def xor_byte_strings(s1, s2)
+        s1.bytes.zip(s2.bytes).map { |(c1,c2)| c1 ^ c2 }.pack('c*')
+      end
+    end
+  end
+end

data/lib/breach_mitigation/railtie.rb

@@ -0,0 +1,28 @@
+require 'breach_mitigation/length_hiding'
+require 'breach_mitigation/masking_secrets'
+
+module BreachMitigation
+  class Railtie < Rails::Railtie
+    initializer "breach-mitigation-rails.insert_middleware" do |app|
+      app.config.middleware.use "BreachMitigation::LengthHiding"
+    end
+  end
+end
+
+# Monkey-patch ActionController::RequestForgeryProtection to use
+# masked CSRF tokens
+module ActionController
+  module RequestForgeryProtection
+    protected
+
+    def verified_request?
+      !protect_against_forgery? || request.get? || request.head? ||
+        BreachMitigation::MaskingSecrets.valid_authenticity_token?(session, params[request_forgery_protection_token]) ||
+        BreachMitigation::MaskingSecrets.valid_authenticity_token?(session, request.headers['X-CSRF-Token'])
+    end
+
+    def form_authenticity_token
+      BreachMitigation::MaskingSecrets.masked_authenticity_token(session)
+    end
+  end
+end

data/lib/breach_mitigation/version.rb

@@ -0,0 +1,3 @@
+module BreachMitigation
+  VERSION = "0.0.1"
+end

metadata

@@ -0,0 +1,91 @@
+--- !ruby/object:Gem::Specification
+name: breach-mitigation-rails
+version: !ruby/object:Gem::Version
+  prerelease: 
+  version: 0.0.1
+platform: ruby
+authors:
+- Bradley Buda
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2013-08-02 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '1.3'
+    none: false
+  name: bundler
+  type: :development
+  prerelease: false
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '1.3'
+    none: false
+- !ruby/object:Gem::Dependency
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+    none: false
+  name: rake
+  type: :development
+  prerelease: false
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+    none: false
+description: Mitigates the BREACH and CRIME attacks on TLS in Rails applications
+email:
+- bradleybuda@gmail.com
+executables: []
+extensions: []
+extra_rdoc_files: []
+files:
+- .gitignore
+- Gemfile
+- LICENSE.txt
+- README.md
+- Rakefile
+- breach-mitigation-rails.gemspec
+- lib/breach-mitigation-rails.rb
+- lib/breach_mitigation/length_hiding.rb
+- lib/breach_mitigation/masking_secrets.rb
+- lib/breach_mitigation/railtie.rb
+- lib/breach_mitigation/version.rb
+homepage: https://github.com/meldium/breach-mitigation-rails
+licenses:
+- MIT
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+  none: false
+required_rubygems_version: !ruby/object:Gem::Requirement
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+  none: false
+requirements: []
+rubyforge_project: 
+rubygems_version: 1.8.23
+signing_key: 
+specification_version: 3
+summary: Uses length-hiding and CSRF token masking to make it more difficult for an
+  attacker to recover plaintext from HTTP responses. See README.md for details.
+test_files: []
+has_rdoc: