diff --git a/cm_crowdin_aosp_sync.py b/cm_crowdin_aosp_sync.py
new file mode 100755
index 0000000..3c40b42
--- /dev/null
+++ b/cm_crowdin_aosp_sync.py
@@ -0,0 +1,277 @@
+#!/usr/bin/python2
+# -*- coding: utf-8 -*-
+# cm_crowdin_sync.py
+#
+# Updates Crowdin source translations and pulls translations
+# directly to CyanogenMod's Git.
+#
+# Copyright (C) 2014 The CyanogenMod Project
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#   http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import codecs
+import git
+import mmap
+import os
+import os.path
+import re
+import shutil
+import subprocess
+import sys
+from urllib import urlretrieve
+from xml.dom import minidom
+
+def purge_caf_additions(strings_base, strings_cm):
+    # Load AOSP file and resources
+    xml_base = minidom.parse(strings_base)
+    list_base_string = xml_base.getElementsByTagName('string')
+    list_base_string_array = xml_base.getElementsByTagName('string-array')
+    list_base_plurals = xml_base.getElementsByTagName('plurals')
+    # Load CM file and resources
+    xml_cm = minidom.parse(strings_cm)
+    list_cm_string = xml_cm.getElementsByTagName('string')
+    list_cm_string_array = xml_cm.getElementsByTagName('string-array')
+    list_cm_plurals = xml_cm.getElementsByTagName('plurals')
+    with codecs.open(strings_cm, 'r', 'utf-8') as f:
+        content = [line.rstrip() for line in f]
+    shutil.copyfile(strings_cm, strings_cm + '.backup')
+    file_this = codecs.open(strings_cm, 'w', 'utf-8')
+
+    # All names from AOSP
+    names_base_string = []
+    names_base_string_array = []
+    names_base_plurals = []
+
+    # Get all names from AOSP
+    for s in list_base_string :
+        names_base_string.append(s.attributes['name'].value)
+    for s in list_base_string_array :
+        names_base_string_array.append(s.attributes['name'].value)
+    for s in list_base_plurals :
+        names_base_plurals.append(s.attributes['name'].value)
+
+    # Get all names from CM
+    content2 = []
+    for s in list_cm_string :
+        name = s.attributes['name'].value
+        if name not in names_base_string:
+            true = 0
+            content2 = []
+            for i in content:
+                if true == 0:
+                    test = re.search('(<string name=\"' + name + ')', i)
+                    if test is not None:
+                        test2 = re.search('(</string>)', i)
+                        if test2:
+                            true = 2
+                        else:
+                            true = 1
+                        i = ''
+                elif true == 1:
+                    test2 = re.search('(</string>)', i)
+                    if test2 is not None:
+                        true = 2
+                    i = ''
+                elif true == 2:
+                    true = 3
+                content2.append(i)
+            content = content2
+    for s in list_cm_string_array :
+        name = s.attributes['name'].value
+        if name not in names_base_string_array:
+            true = 0
+            content2 = []
+            for i in content:
+                if true == 0:
+                    test = re.search('(<string-array name=\"' + name + ')', i)
+                    if test is not None:
+                        test2 = re.search('(</string-array>)', i)
+                        if test2:
+                            true = 2
+                        else:
+                            true = 1
+                        i = ''
+                elif true == 1:
+                    test2 = re.search('(</string-array>)', i)
+                    if test2 is not None:
+                        true = 2
+                    i = ''
+                elif true == 2:
+                    true = 3
+                content2.append(i)
+            content = content2
+    for s in list_cm_plurals :
+        name = s.attributes['name'].value
+        if name not in names_base_plurals:
+            true = 0
+            content2 = []
+            for i in content:
+                if true == 0:
+                    test = re.search('(<plurals name=\"' + name + ')', i)
+                    if test is not None:
+                        test2 = re.search('(</plurals>)', i)
+                        if test2:
+                            true = 2
+                        else:
+                            true = 1
+                        i = ''
+                elif true == 1:
+                    test2 = re.search('(</plurals>)', i)
+                    if test2 is not None:
+                        true = 2
+                    i = ''
+                elif true == 2:
+                    # The actual purging is done!
+                    true = 3
+                content2.append(i)
+            content = content2
+
+    for addition in content:
+        file_this.write(addition + '\n')
+    file_this.close()
+
+def push_as_commit(path, name, branch):
+    # CM gerrit nickname
+    username = 'your_nickname'
+
+    # Get path
+    path = os.getcwd() + '/' + path
+
+    # Create git commit
+    repo = git.Repo(path)
+    repo.git.add(path)
+    try:
+        repo.git.commit(m='Automatic translation import')
+        repo.git.push('ssh://' + username + '@review.cyanogenmod.org:29418/' + name, 'HEAD:refs/for/' + branch)
+        print 'Succesfully pushed commit for ' + name
+    except:
+        # If git commit fails, it's probably because of no changes.
+        # Just continue.
+        print 'No commit pushed (probably empty?) for ' + name
+
+print('Welcome to the CM Crowdin sync script!')
+
+print('\nSTEP 0: Checking dependencies')
+# Check for Ruby version of crowdin-cli
+if subprocess.check_output(['rvm', 'all', 'do', 'gem', 'list', 'crowdin-cli', '-i']) == 'true':
+    sys.exit('You have not installed crowdin-cli. Terminating.')
+else:
+    print('Found: crowdin-cli')
+# Check for caf.xml
+if not os.path.isfile('caf.xml'):
+    sys.exit('You have no caf.xml. Terminating.')
+else:
+    print('Found: caf.xml')
+# Check for android/default.xml
+if not os.path.isfile('android/default.xml'):
+    sys.exit('You have no android/default.xml. Terminating.')
+else:
+    print('Found: android/default.xml')
+# Check for extra_packages.xml
+if not os.path.isfile('extra_packages.xml'):
+    sys.exit('You have no extra_packages.xml. Terminating.')
+else:
+    print('Found: extra_packages.xml')
+# Check for repo
+try:
+    subprocess.check_output(['which', 'repo'])
+except:
+    sys.exit('You have not installed repo. Terminating.')
+
+print('\nSTEP 1: Removing CAF additions')
+# Load caf.xml
+print('Loading caf.xml')
+xml = minidom.parse('caf.xml')
+items = xml.getElementsByTagName('item')
+
+# Store all created cm_caf.xml files in here.
+# Easier to remove them afterwards, as they cannot be committed
+cm_caf = []
+
+for item in items:
+    # Create tmp dir for download of AOSP base file
+    path_to_values = item.attributes["path"].value
+    subprocess.call(['mkdir', '-p', 'tmp/' + path_to_values])
+    for aosp_item in item.getElementsByTagName('aosp'):
+        url = aosp_item.firstChild.nodeValue
+        xml_file = aosp_item.attributes["file"].value
+        path_to_base = 'tmp/' + path_to_values + '/' + xml_file
+        path_to_cm = path_to_values + '/' + xml_file
+        urlretrieve(url, path_to_base)
+        purge_caf_additions(path_to_base, path_to_cm)
+        cm_caf.append(path_to_cm)
+        print('Purged ' + path_to_cm + ' from CAF additions')
+
+print('\nSTEP 2: Upload Crowdin source translations')
+# Execute 'crowdin-cli upload sources' and show output
+print(subprocess.check_output(['crowdin-cli', '-c', 'crowdin-aosp.yaml', 'upload', 'sources']))
+
+print('\nSTEP 3: Download Crowdin translations')
+# Execute 'crowdin-cli download' and show output
+print(subprocess.check_output(['crowdin-cli', '-c', 'crowdin-aosp.yaml', "download"]))
+
+print('\nSTEP 4A: Revert purges')
+for purged_file in cm_caf:
+    os.remove(purged_file)
+    shutil.move(purged_file + '.backup', purged_file)
+    print('Reverted purged file ' + purged_file)
+
+print('\nSTEP 4B: Clean up of temp dir')
+# We are done with cm_caf.xml files, so remove tmp/
+shutil.rmtree(os.getcwd() + '/tmp')
+
+print('\nSTEP 4C: Clean up of empty translations')
+# Some line of code that I found to find all XML files
+result = [os.path.join(dp, f) for dp, dn, filenames in os.walk(os.getcwd()) for f in filenames if os.path.splitext(f)[1] == '.xml']
+for xml_file in result:
+    # We hate empty, useless files. Crowdin exports them with <resources/> (sometimes with xliff).
+    # That means: easy to find
+    if '<resources/>' in open(xml_file).read():
+        print ('Removing ' + xml_file)
+        os.remove(xml_file)
+    elif '<resources xmlns:xliff="urn:oasis:names:tc:xliff:document:1.2"/>' in open(xml_file).read():
+        print ('Removing ' + xml_file)
+        os.remove(xml_file)
+
+print('\nSTEP 5: Push translations to Git')
+# Get all files that Crowdin pushed
+proc = subprocess.Popen(['crowdin-cli', '-c', 'crowdin-aosp.yaml', 'list', 'sources'],stdout=subprocess.PIPE)
+xml = minidom.parse('android/default.xml')
+xml_extra = minidom.parse('extra_packages.xml')
+items = xml.getElementsByTagName('project')
+items += xml_extra.getElementsByTagName('project')
+all_projects = []
+
+for path in iter(proc.stdout.readline,''):
+    # Remove the \n at the end of each line
+    path = path.rstrip()
+    # Get project root dir from Crowdin's output
+    m = re.search('/(.*Superuser)/Superuser.*|/(.*LatinIME).*|/(frameworks/base).*|/(.*CMFileManager).*|/(device/.*/.*)/.*/res/values.*|/(hardware/.*/.*)/.*/res/values.*|/(.*)/res/values.*', path)
+    for good_path in m.groups():
+        # When a project has multiple translatable files, Crowdin will give duplicates.
+        # We don't want that (useless empty commits), so we save each project in all_projects
+        # and check if it's already in there.
+        if good_path is not None and not good_path in all_projects:
+            all_projects.append(good_path)
+            for project_item in items:
+                # We need to have the Github repository for the git push url.
+                # Obtain them from android/default.xml or extra_packages.xml.
+                if project_item.attributes["path"].value == good_path:
+                    if project_item.hasAttribute('revision'):
+                        branch = project_item.attributes['revision'].value
+                    else:
+                        branch = 'cm-11.0'
+                    print 'Committing ' + project_item.attributes['name'].value + ' on branch ' + branch + ' (based on android/default.xml or extra_packages.xml)'
+                    push_as_commit(good_path, project_item.attributes['name'].value, branch)
+
+print('\nSTEP 6: Done!')