# SpamAssassin rules file: header tests
#
# Please don't modify this file as your changes will be overwritten with
# the next update. Use @@LOCAL_RULES_DIR@@/local.cf instead.
# See 'perldoc Mail::SpamAssassin::Conf' for details.
#
# <@LICENSE>
# Copyright 2004 Apache Software Foundation
#
# 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.
# </@LICENSE>
#
###########################################################################
require_version @@VERSION@@
header HEAD_LONG eval:check_for_long_header()
describe HEAD_LONG Message headers are very long
header NO_REAL_NAME From =~ /^["\s]*\<?\S+\@\S+\>?\s*$/
describe NO_REAL_NAME From: does not include a real name
header FROM_ENDS_IN_NUMS From =~ /\d\d\@/
describe FROM_ENDS_IN_NUMS From: ends in numbers
header FROM_STARTS_WITH_NUMS From =~ /^\d\d/
describe FROM_STARTS_WITH_NUMS From: starts with nums
# note: anchored for speed
header FROM_HAS_MIXED_NUMS From =~ /(?:^|\D)\d+[a-z]+\d+\S*\@/i
describe FROM_HAS_MIXED_NUMS From: contains numbers mixed in with letters
# idea from Robert Menschel <RMSA@Menschel.net>
header FROM_HAS_MIXED_NUMS3 From:addr =~ /^[a-z]+\d+[a-z]+\d+[a-z]+\w*\@/i
describe FROM_HAS_MIXED_NUMS3 From: contains numbers mixed in with letters
# Faked addresses tend to come from big public sites. Stats show that
# 5 digits is enough to get a 1.0 s/o ratio; 4 is too low (probably due
# to folks called "jmason2002@yahoo.com" for example).
header ADDR_NUMS_AT_BIGSITE ALL =~ /^(To|From|Cc|Reply-To):\s*<?\S+\d{5,}\@(?:aol|bigfoot|compuserve|excite|hotmail|juno|prodigy|yahoo)\.(?:com|net|org)/mi
describe ADDR_NUMS_AT_BIGSITE Uses an address with lots of numbers, at a big ISP
header __FROM_JUST_NUMBER From:addr =~ /^\d+\@/
header __FROM_PHONE From:addr =~ /^\d{3}(?:[-.]?\d{3}[-.]?\d{4}|\d{7})\@/
meta FROM_ALL_NUMS (__FROM_JUST_NUMBER && !__FROM_PHONE)
describe FROM_ALL_NUMS From an address that is all numbers (non-phone)
header FROM_OFFERS From:addr =~ /\@\S*offers(?![eo]n\b)/i
describe FROM_OFFERS From address is "at something-offers"
header FROM_NO_USER From =~ /(?:^\@|<\@| \@[^\)<]*$|<>)/ [if-unset: unset@unset.unset]
describe FROM_NO_USER From: has no local-part before @ sign
header TO_NO_USER To =~ /(?:^\@|<\@| \@[^\)<]*$|<>)/ [if-unset: unset@unset.unset]
describe TO_NO_USER To: has no local-part before @ sign
header TO_EMPTY To =~ /^\s*$/ [if-unset: UNSET]
describe TO_EMPTY To: is empty
header REPLY_TO_EMPTY Reply-To =~ /^\s*$/ [if-unset: UNSET]
describe REPLY_TO_EMPTY Reply-To: is empty
header TO_ADDRESS_EQ_REAL To =~ /^\s*"([^"@]+\@[^"@]+)"\s+<\1>\s*$/i
describe TO_ADDRESS_EQ_REAL To: repeats address as real name
# NOTE: this is what 100% valid undisclosed-recipients mails look like.
# If this gets a high score, that's a bug!
header UNDISC_RECIPS To =~ /^undisclosed-recipients?:\s*;$/
describe UNDISC_RECIPS Valid-looking To "undisclosed-recipients"
# also 100% valid
header FAKED_UNDISC_RECIPS To =~ /undisclosed[_ ]*recipient(?:s[^:]|[^s])/i
describe FAKED_UNDISC_RECIPS Faked To "Undisclosed-Recipients"
header PLING_QUERY Subject =~ /\?.*!|!.*\?/
describe PLING_QUERY Subject has exclamation mark and question mark
header SUBJ_HAS_UNIQ_ID eval:check_for_unique_subject_id()
describe SUBJ_HAS_UNIQ_ID Subject contains a unique ID
header SUBJ_HAS_SPACES Subject =~ /(?:\s{6}|\t\s|\s\t)\S/
describe SUBJ_HAS_SPACES Subject contains lots of white space
header SUBJ_ALL_CAPS eval:subject_is_all_caps()
describe SUBJ_ALL_CAPS Subject is all capitals
header MSGID_SPAM_99X9XX99 MESSAGEID =~ /^<\d\d\d\d\d\d[a-z]\d[a-z][a-z]\d\d\$[a-z][a-z][a-z]\d\d\d\d\d\$\d\d\d\d\d\d\d\d\@/
describe MSGID_SPAM_99X9XX99 Spam tool Message-Id: (99x9xx99 variant)
header MSGID_SPAM_ALPHA_NUM MESSAGEID =~ /<[A-Z]{7}-000[0-9]{10}\@[a-z]*>/
describe MSGID_SPAM_ALPHA_NUM Spam tool Message-Id: (alpha-numeric variant)
header MSGID_SPAM_CAPS Message-ID =~ /^\s*<?[A-Z]+\@(?!(?:mailcity|whowhere)\.com)/
describe MSGID_SPAM_CAPS Spam tool Message-Id: (caps variant)
header MSGID_SPAM_LETTERS Message-Id =~ /<[a-z]{5,}\@(\S+\.)+\S+>/
describe MSGID_SPAM_LETTERS Spam tool Message-Id: (letters variant)
header MSGID_SPAM_ZEROES MESSAGEID =~ /<0000[0-9a-f]{8}\$0000[0-9a-f]{4}\$0000[0-9a-f]{4}\@/
describe MSGID_SPAM_ZEROES Spam tool Message-Id: (12-zeroes variant)
header MSGID_NO_HOST MESSAGEID =~ /\@>(?:$|\s)/m
describe MSGID_NO_HOST Message-Id has no hostname
header MSGID_OUTLOOK_INVALID eval:check_outlook_message_id()
describe MSGID_OUTLOOK_INVALID Message-Id is fake (in Outlook Express format)
# catches a few spams missed by MSGID_OUTLOOK_INVALID
meta MSGID_DOLLARS (__OUTLOOK_DOLLARS_MSGID && !__HAS_OUTLOOK_IN_MAILER && !__UNUSABLE_MSGID)
describe MSGID_DOLLARS Message-Id has pattern used in spam
# bit of a ratware rule, but catches a bit more than just the one ratware
header __MSGID_RANDY Message-ID =~ /<[a-z\d][a-z\d\$-]{10,29}[a-z\d]\@[a-z\d][a-z\d.]{3,12}[a-z\d]>/
# heuristic to eliminate most good Message-ID formats
header __MSGID_OK_HEX Message-ID =~ /\b[a-f\d]{8}\b/
header __MSGID_OK_DIGITS Message-ID =~ /\d{10}/
header __MSGID_OK_HOST Message-ID =~ /\@(?:\D{2,}|(?:\d{1,3}\.){3}\d{1,3})>/
meta MSGID_RANDY (__MSGID_RANDY && !(__MSGID_OK_HEX || __MSGID_OK_HOST || __MSGID_OK_HOST))
describe MSGID_RANDY Message-Id has pattern used in spam
# bug 3395
header MSGID_YAHOO_CAPS Message-ID =~ /<[A-Z]+\@yahoo.com>/
describe MSGID_YAHOO_CAPS Message-ID has ALLCAPS@yahoo.com
###########################################################################
header __MSGID_BEFORE_RECEIVED ALL =~ /\nMessage-Id:.*\nReceived:/si
header __MSGID_BEFORE_OKAY Message-Id =~ /\@[a-z0-9.-]+\.(?:yahoo|wanadoo)(?:\.[a-z]{2,3}){1,2}>/
meta MSGID_FROM_MTA_HEADER (__MSGID_BEFORE_RECEIVED && !__MSGID_BEFORE_OKAY)
describe MSGID_FROM_MTA_HEADER Message-Id was added by a relay
header MSGID_FROM_MTA_ID eval:message_id_from_mta()
describe MSGID_FROM_MTA_ID Message-Id for external message added locally
header MSGID_FROM_MTA_HOTMAIL Message-Id =~ /<MC\d{1,2}-F{1,2}\w{21,22}\@\S*hotmail\.com>/
describe MSGID_FROM_MTA_HOTMAIL Message-Id was added by a hotmail.com relay
###########################################################################
header DATE_SPAMWARE_Y2K Date =~ /^[A-Z][a-z]{2}, \d\d [A-Z][a-z]{2} [0-6]\d \d\d:\d\d:\d\d [A-Z]{3}$/
describe DATE_SPAMWARE_Y2K Date header uses unusual Y2K formatting
header INVALID_DATE Date !~ /^\s*(?:(?:Mon|Tue|Wed|Thu|Fri|Sat|Sun), )?[0-3 ]?[0-9] (?:Jan|Feb|Ma[ry]|Apr|Ju[nl]|Aug|Sep|Oct|Nov|Dec) (?:[12][901])?[0-9]{2} [0-2][0-9](?:\:[0-5][0-9]){1,2} (?:[+-][0-9]{4}|UT|[A-Z]{2,3}T)(?:\s+\(.*\))?\s*$/ [if-unset: Wed, 31 Jul 2002 16:41:57 +0200]
describe INVALID_DATE Invalid Date: header (not RFC 2822)
# allow +1300, NZ timezone
header INVALID_DATE_TZ_ABSURD Date =~ /[-+](?:1[4-9]\d\d|[2-9]\d\d\d)$/
describe INVALID_DATE_TZ_ABSURD Invalid Date: header (timezone does not exist)
header INVALID_TZ_CST ALL =~ /[+-]\d\d[30]0(?<!-0600|-0500|\+0800|\+0930|\+1030)\s+(?:\bCST\b|\(CST\))/
describe INVALID_TZ_CST Invalid date in header (wrong CST timezone)
header INVALID_TZ_EST ALL =~ /[+-]\d\d[30]0(?<!-0500|-0300|\+1000|\+1100)\s+(?:\bEST\b|\(EST\))/
describe INVALID_TZ_EST Invalid date in header (wrong EST timezone)
header INVALID_TZ_GMT ALL =~ /[+-]\d\d[30]0(?<![+-]0000)\s+(?:\b(?:GMT|UTC)\b(?![\w+-])|\((?:GMT|UTC)\))/
describe INVALID_TZ_GMT Invalid date in header (wrong GMT/UTC timezone)
header DATE_IN_PAST_03_06 eval:check_for_shifted_date('-6', '-3')
describe DATE_IN_PAST_03_06 Date: is 3 to 6 hours before Received: date
header DATE_IN_PAST_06_12 eval:check_for_shifted_date('-12', '-6')
describe DATE_IN_PAST_06_12 Date: is 6 to 12 hours before Received: date
header DATE_IN_PAST_12_24 eval:check_for_shifted_date('-24', '-12')
describe DATE_IN_PAST_12_24 Date: is 12 to 24 hours before Received: date
header DATE_IN_PAST_24_48 eval:check_for_shifted_date('-48', '-24')
describe DATE_IN_PAST_24_48 Date: is 24 to 48 hours before Received: date
header DATE_IN_PAST_48_96 eval:check_for_shifted_date('-96', '-48')
describe DATE_IN_PAST_48_96 Date: is 48 to 96 hours before Received: date
header DATE_IN_PAST_96_XX eval:check_for_shifted_date('undef', '-96')
describe DATE_IN_PAST_96_XX Date: is 96 hours or more before Received: date
header DATE_IN_FUTURE_03_06 eval:check_for_shifted_date('3', '6')
describe DATE_IN_FUTURE_03_06 Date: is 3 to 6 hours after Received: date
header DATE_IN_FUTURE_06_12 eval:check_for_shifted_date('6', '12')
describe DATE_IN_FUTURE_06_12 Date: is 6 to 12 hours after Received: date
header DATE_IN_FUTURE_12_24 eval:check_for_shifted_date('12', '24')
describe DATE_IN_FUTURE_12_24 Date: is 12 to 24 hours after Received: date
header DATE_IN_FUTURE_24_48 eval:check_for_shifted_date('24', '48')
describe DATE_IN_FUTURE_24_48 Date: is 24 to 48 hours after Received: date
header DATE_IN_FUTURE_48_96 eval:check_for_shifted_date('48', '96')
describe DATE_IN_FUTURE_48_96 Date: is 48 to 96 hours after Received: date
header DATE_IN_FUTURE_96_XX eval:check_for_shifted_date('96', 'undef')
describe DATE_IN_FUTURE_96_XX Date: is 96 hours or more after Received: date
header UNRESOLVED_TEMPLATE ALL =~ /^(?!(?i:X-UIDL|X-Face|To|Cc|From|Subject|References|In-Reply-To|(?:X-|Resent-|X-Original-)?Message-Id):)[\w-]{1,24}:(?:[^\n]{0,100}|\n[ \t]){0,2}%[A-Z][A-Z_-]/m
describe UNRESOLVED_TEMPLATE Headers contain an unresolved template
###########################################################################
# illegal characters that should be MIME encoded
# might want to exempt users using languages that don't use Latin
# alphabets, but do it in the eval
header SUBJ_ILLEGAL_CHARS eval:check_illegal_chars('Subject','0.00','2')
describe SUBJ_ILLEGAL_CHARS Subject contains too many raw illegal characters
header FROM_ILLEGAL_CHARS eval:check_illegal_chars('From','0.20','2')
describe FROM_ILLEGAL_CHARS From contains too many raw illegal characters
header HEAD_ILLEGAL_CHARS eval:check_illegal_chars('ALL','0.005','2')
describe HEAD_ILLEGAL_CHARS Header contains too many raw illegal characters
###########################################################################
# ADV tags in various languages
header ENGLISH_UCE_SUBJECT Subject =~ /^[^0-9a-z]*adv(?:ert)?\b/i
describe ENGLISH_UCE_SUBJECT Subject contains an English UCE tag
# alan premselaar <alien@12inch.com>, see SpamAssassin-talk list 2003-03
# quinlan: 2003-03-23 here are more generic Japanese iso-2022-jp codes
# ("not yet acceptance" or "email") + "announcement"
# FWIW, according to Peter Evans, this should be sufficient to catch the
# UCE tag and a common attempt at evasion (using the "sue" instead of
# "mi" Chinese character).
header JAPANESE_UCE_SUBJECT Subject =~ /\e\$B.*(?:L\$>5Bz|EE;R%a!<%k)9-9p/
describe JAPANESE_UCE_SUBJECT Subject contains a Japanese UCE tag
# quinlan: "advertisement" in Russian KOI8-R
# (no longer common, but worth noting in future)
#header RUSSIAN_UCE_SUBJECT Subject =~ /\xf0\xe5\xea\xeb\xe0\xec\xf3/
#describe RUSSIAN_UCE_SUBJECT Subject contains a Russian UCE tag
# Korean UCE Subject: lines are usually 8-bit, but are occasionally encoded
# with quoted-printable or base64.
#
# \xbc\xba\xc0\xce means "adult"
# \xb1\xa4\xb0\xed means "advertisement"
# \xc1\xa4\xba\xb8 means "information"
# \xc8\xab\xba\xb8 means "publicity"
#
# Each two byte sequence is one Korean letter; the spaces and periods are
# sometimes used to obscure the words. \xb1\xa4\xb0\xed is the most common
# tag and is sometimes very obscured so we look harder.
#
header KOREAN_UCE_SUBJECT Subject =~ /[({[<][. ]*(?:\xbc\xba[. ]*\xc0\xce[. ]*)?(?:\xb1\xa4(?:[. ]*|[\x00-\x7f]{0,3})\xb0\xed|\xc1\xa4[. ]*\xba\xb8|\xc8\xab[. ]*\xba\xb8)[. ]*[)}\]>]/
describe KOREAN_UCE_SUBJECT Subject: contains Korean unsolicited email tag
###########################################################################
header FROM_AND_TO_SAME eval:check_for_from_to_same()
describe FROM_AND_TO_SAME From and To are the same, but not exactly
header FORGED_RCVD_HELO eval:check_for_forged_received_helo()
describe FORGED_RCVD_HELO Received: contains a forged HELO
header RCVD_HELO_IP_MISMATCH eval:helo_ip_mismatch()
describe RCVD_HELO_IP_MISMATCH Received: HELO and IP do not match, but should
header RCVD_NUMERIC_HELO eval:check_for_numeric_helo()
describe RCVD_NUMERIC_HELO Received: contains an IP address used for HELO
header RCVD_ILLEGAL_IP eval:check_for_illegal_ip()
describe RCVD_ILLEGAL_IP Received: contains illegal IP address
# no legit mailer claims that their mailserver has no name
# overlaps with RCVD_DOUBLE_IP*, but let's see how it is scored
header RCVD_BY_IP Received =~ /\bby\s+\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}(?<!127\.0\.0\.1)\b/
describe RCVD_BY_IP Received by mail server with no name
# two reliable signatures
header __DOUBLE_IP_SPAM_1 Received =~ /from \[\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}\] by \d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3} with/
header __DOUBLE_IP_SPAM_2 Received =~ /from\s+\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3}\s+by\s+\d{1,3}\.\d{1,3}\.\d{1,3}\.\d{1,3};/
# loose match
header __DOUBLE_IP_LOOSE Received =~ /(?:\b(?:from|by)\b.{1,4}\b\d{1,3}[._-]\d{1,3}[._-]\d{1,3}[._-]\d{1,3}(?<!127\.0\.0\.1)\b.{0,4}){2}/i
# spam signature
meta RCVD_DOUBLE_IP_SPAM (__DOUBLE_IP_SPAM_1 || __DOUBLE_IP_SPAM_2)
describe RCVD_DOUBLE_IP_SPAM Bulk email fingerprint (double IP) found
# other matches
meta RCVD_DOUBLE_IP_LOOSE (__DOUBLE_IP_LOOSE && !RCVD_DOUBLE_IP_SPAM)
describe RCVD_DOUBLE_IP_LOOSE Received: by and from look like IP addresses
header FORGED_AOL_RCVD eval:check_for_fake_aol_relay_in_rcvd()
describe FORGED_AOL_RCVD Received forged, contains fake AOL relays
header FORGED_TELESP_RCVD Received =~ /\.(?!br).. \(\d+-\d+-\d+-\d+\.dsl\.telesp\.net\.br /
describe FORGED_TELESP_RCVD Contains forged hostname for a DSL IP in Brazil
# a forged Hotmail message; host HELO'd as hotmail.com, but it wasn't
header FORGED_HOTMAIL_RCVD eval:check_for_forged_hotmail_received_headers()
describe FORGED_HOTMAIL_RCVD Forged hotmail.com 'Received:' header found
# this, by comparison is more common: from was @hotmail.com, but it wasn't
header FORGED_HOTMAIL_RCVD2 eval:check_for_no_hotmail_received_headers()
describe FORGED_HOTMAIL_RCVD2 hotmail.com 'From' address, but no 'Received:'
header FORGED_EUDORAMAIL_RCVD eval:check_for_forged_eudoramail_received_headers()
describe FORGED_EUDORAMAIL_RCVD Forged eudoramail.com 'Received:' header found
header FORGED_YAHOO_RCVD eval:check_for_forged_yahoo_received_headers()
describe FORGED_YAHOO_RCVD 'From' yahoo.com does not match 'Received' headers
header FORGED_JUNO_RCVD eval:check_for_forged_juno_received_headers()
describe FORGED_JUNO_RCVD 'From' juno.com does not match 'Received' headers
header FORGED_GW05_RCVD eval:check_for_forged_gw05_received_headers()
describe FORGED_GW05_RCVD Forged 'by gw05' 'Received:' header found
# not used directly right now due to FPs; but CONFIRMED_FORGED turns it
# into a 1.0 S/O rule anyway, so that's not a problem ;)
# 2.626 3.6340 1.5251 0.704 0.34 1.44 FORGED_RCVD_TRAIL
# 0.956 3.3890 0.0000 1.000 0.98 4.30 CONFIRMED_FORGED
header __FORGED_RCVD_TRAIL eval:check_for_forged_received_trail()
# forgery meta-rules: more reliable than their inputs
meta CONFIRMED_FORGED (__FORGED_RCVD_TRAIL && (FORGED_AOL_RCVD || FORGED_HOTMAIL_RCVD || FORGED_EUDORAMAIL_RCVD || FORGED_YAHOO_RCVD || FORGED_JUNO_RCVD || FORGED_GW05_RCVD || FORGED_MX_HOTMAIL))
describe CONFIRMED_FORGED Received headers are forged
meta MULTI_FORGED ((FORGED_AOL_RCVD + FORGED_HOTMAIL_RCVD + FORGED_EUDORAMAIL_RCVD + FORGED_YAHOO_RCVD + FORGED_JUNO_RCVD + FORGED_GW05_RCVD) > 1)
describe MULTI_FORGED Received headers indicate multiple forgeries
header NONEXISTENT_CHARSET Content-Type =~ /charset=.?DEFAULT/
describe NONEXISTENT_CHARSET Character set doesn't exist
header CHARSET_FARAWAY_HEADER eval:check_for_faraway_charset_in_headers()
describe CHARSET_FARAWAY_HEADER A foreign language charset used in headers
tflags CHARSET_FARAWAY_HEADER userconf
header X_PRIORITY_HIGH X-Priority =~ /^1/
describe X_PRIORITY_HIGH Sent with 'X-Priority' set to high
header X_MSMAIL_PRIORITY_HIGH X-Msmail-Priority =~ /^High/
describe X_MSMAIL_PRIORITY_HIGH Sent with 'X-Msmail-Priority' set to high
# this variant is local, using the Received hdr itself...
header ROUND_THE_WORLD_LOCAL eval:check_for_round_the_world_received_helo()
describe ROUND_THE_WORLD_LOCAL Received: says mail sent around the world (HELO)
# and this one uses a DNS reverse lookup. so now we can use a version
# of this test without a net connection, or in mass-check etc.
header ROUND_THE_WORLD eval:check_for_round_the_world_received_revdns()
describe ROUND_THE_WORLD Received: says mail sent around the world (DNS)
tflags ROUND_THE_WORLD net
header MISSING_DATE Date =~ /^UNSET$/ [if-unset: UNSET]
describe MISSING_DATE Missing Date: header
# this is a quite common false positive, as it's legal to remove a To but leave
# a CC. so don't score it high.
header MISSING_HEADERS eval:check_for_missing_to_header()
describe MISSING_HEADERS Missing To: header
header __HAS_SUBJECT exists:Subject
meta MISSING_SUBJECT !__HAS_SUBJECT
describe MISSING_SUBJECT Missing Subject: header
header SUSPICIOUS_RECIPS eval:similar_recipients('0.65','undef')
describe SUSPICIOUS_RECIPS Similar addresses in recipient list
header SORTED_RECIPS eval:sorted_recipients()
describe SORTED_RECIPS Recipient list is sorted by address
header GAPPY_SUBJECT Subject =~ /\b(?:[a-z]([-_. =~\/:,*!\@\#\$\%\^&+;\"\'<>\\])\1{0,2}){4,}/i
describe GAPPY_SUBJECT Subject: contains G.a.p.p.y-T.e.x.t
### header existence tests (description is added automatically)
# X-Fix example: NTMail fixed non RFC822 compliant EMail message
#
# X-PMFLAGS is all caps
#
# Headers that seem to only be used by a single spamming software and
# are found together in the same message:
# 1. X-MailingID and X-ServerHost
# 2. X-Stormpost-To and X-List-Unsubscribe
#
# not spammish: X-EM-Registration, X-EM-Version, X-Antiabuse, X-List-Host,
# X-Message-Id
# bad FP rate: Comment, Date-warning
header X_LIBRARY exists:X-Library
describe X_LIBRARY Message has X-Library header
header __HAS_MIMEOLE exists:X-MimeOLE
header __HAS_MSMAIL_PRI exists:X-MSMail-Priority
header __HAS_SQUIRRELMAIL_IN_MAILER X-Mailer =~ /SquirrelMail\b/
meta MISSING_MIMEOLE (__HAS_MSMAIL_PRI && !__HAS_MIMEOLE && !__HAS_SQUIRRELMAIL_IN_MAILER)
describe MISSING_MIMEOLE Message has X-MSMail-Priority, but no X-MimeOLE
header __HAS_X_MAILER exists:X-Mailer
header __HAS_OUTLOOK_IN_MAILER X-Mailer =~ /Microsoft (CDO|Outlook|Office Outlook)\b/
header __IS_EXCH X-MimeOLE =~ /Produced By Microsoft Exchange V/
header __HAS_X_PRIORITY exists:X-Priority
header __USER_AGENT exists:User-Agent
meta PRIORITY_NO_NAME ((__HAS_X_PRIORITY || __HAS_MSMAIL_PRI) && !__HAS_X_MAILER && !__IS_EXCH && !__USER_AGENT)
describe PRIORITY_NO_NAME Message has priority, but no X-Mailer/User-Agent
header SUBJ_AS_SEEN Subject =~ /\bAs Seen/i
describe SUBJ_AS_SEEN Subject contains "As Seen"
header SUBJ_DOLLARS Subject =~ /^\$[0-9.,]+\b/
describe SUBJ_DOLLARS Subject starts with dollar amount
header SUBJ_FOR_ONLY Subject =~ /For Only/i
describe SUBJ_FOR_ONLY Subject contains "For Only"
header SUBJ_FREE_CAP Subject =~ /FRE{2,}|F.R.E.E\b/
describe SUBJ_FREE_CAP Subject contains "FREE" in CAPS
header SUB_FREE_OFFER Subject =~ /^fre{2,}\b/i
describe SUB_FREE_OFFER Subject starts with "Free"
header SUBJ_GUARANTEED Subject =~ /^guaranteed|(?-i:GUARANTEE)/i
describe SUBJ_GUARANTEED Subject GUARANTEED
header SUB_HELLO Subject =~ /^hello\b/i
describe SUB_HELLO Subject starts with "Hello"
header SUBJ_LIFE_INSURANCE Subject =~ /life\s+insurance/i
describe SUBJ_LIFE_INSURANCE Subject includes "life insurance"
header SUBJ_YOUR_DEBT Subject =~ /Your (?:Bills|Debt|Credit)/i
describe SUBJ_YOUR_DEBT Subject contains "Your Bills" or similar
header SUBJ_YOUR_FAMILY Subject =~ /Your Family/i
describe SUBJ_YOUR_FAMILY Subject contains "Your Family"
header SUBJ_YOUR_OWN Subject =~ /Your Own/i
describe SUBJ_YOUR_OWN Subject contains "Your Own"
# the real services never HELO as 'foo.com', instead 'mail.foo.com' or
# something like that. Note: be careful when expanding this... legit dotcom
# HELOers include: hotmail.com, drizzle.com, lockergnome.com.
header RCVD_FAKE_HELO_DOTCOM Received =~ /^from (?:msn|yahoo|yourwebsite|lycos|excite|cs|aol|localhost|koreanmail|allexecs|mydomain|juno|eudoramail|compuserve|desertmail|excite|caramail)\.com \(/m
describe RCVD_FAKE_HELO_DOTCOM Received contains a faked HELO hostname
header ADDRESS_IN_SUBJECT eval:check_for_to_in_subject('address')
describe ADDRESS_IN_SUBJECT To: address appears in Subject
header SUBJECT_DIET Subject =~ /\bLose .*(?:pounds|lbs|weight)/i
describe SUBJECT_DIET Subject talks about losing pounds
header EXTRA_MPART_TYPE Content-Type =~ /(?:\s*multipart\/)?.* type=/i
describe EXTRA_MPART_TYPE Header has extraneous Content-type:...type= entry
header TO_RECIP_MARKER To =~ /\#recipient\#/
describe TO_RECIP_MARKER To header contains 'recipient' marker
# MIME boundary tests; spam tools use distinctive patterns.
header MIME_BOUND_DD_DIGITS Content-Type =~ /boundary=\"--\d+\"/
describe MIME_BOUND_DD_DIGITS Spam tool pattern in MIME boundary
header MIME_BOUND_DIGITS_7 Content-Type =~ /boundary=\d{9}\.\d{13}/
describe MIME_BOUND_DIGITS_7 Spam tool pattern in MIME boundary
header MIME_BOUND_DIGITS_15 Content-Type =~ /boundary=\"\d{15,}\"/
describe MIME_BOUND_DIGITS_15 Spam tool pattern in MIME boundary
header MIME_BOUND_MANY_HEX Content-Type =~ /boundary="[\da-f]{8}(?:-[\da-f]{4}){3}-[\da-f]{12}"/
describe MIME_BOUND_MANY_HEX Spam tool pattern in MIME boundary
header __NEXTPART_ALL Content-Type =~ /NextPart/
header __NEXTPART_NORMAL Content-Type =~ /="(?:----_?=_)?NextPart_[\dA-F]{3}(_[\dA-F]{3,8})?_[\dA-F]{8}\.[\dA-F]{8}"/
meta MIME_BOUND_NEXTPART (__NEXTPART_ALL && !__NEXTPART_NORMAL)
describe MIME_BOUND_NEXTPART Spam tool pattern in MIME boundary
header MIME_BOUND_RKFINDY Content-Type =~ /boundary=\"=_NextPart_2rfkindysadvnqw3nerasdf\"/
describe MIME_BOUND_RKFINDY Spam tool pattern in MIME boundary (rfkindy)
# note: the first alternation is anchored for speed
header TO_MALFORMED To !~ /(?:^|[^\S"])(?:(?:\"[^\"]+\"|\S+)\@\S+\.\S+|^\s*.+:\s*;|^\s*\"[^\"]+\":\s*;|^\s*\([^\)]*\)\s*$|<\S+(?:\!\S+){1,}>|^\s*$)/ [if-unset: unset@unset.unset]
describe TO_MALFORMED To: has a malformed address
# Most/all of these require that From addresses do not start with numbers.
header FROM_NUM_AT_WEBMAIL From:addr =~ /^\d\S+\@(?:msn\.com|flashmail\.com|mailexcite\.com|prodigy\.net|yahoo\.\S+|hotmail\.com|eudoramail\.com|aol\.com|excite\.com|email\.com|earthlink\.net|geocities\.com|hknetmail\.com|angelfire\.com)/i
describe FROM_NUM_AT_WEBMAIL From address is webmail, but starts with a number
header FROM_WEBMAIL_END_NUMS6 From:addr =~ /\d\d\d\d\d\d\@(?:aol|msn|bigfoot|compuserve|excite|hotmail|juno|prodigy|yahoo)\.(?:com|net|org)/i
describe FROM_WEBMAIL_END_NUMS6 From webmail service and address ends in numbers
header ADDR_FREE From =~ /\b(?-i:F)ree(?-i:[ A-Z]).*</i
describe ADDR_FREE From Address contains FREE
# common spam-dropping: To: C:\VICTIMS.txt@yourmx.org
header TO_TXT To =~ /\.txt[\'\"]?\@/i
describe TO_TXT Sent to a text file
header CHINA_HEADER ALL =~ /\@china\.com/i
describe CHINA_HEADER Involves 'china.com'
header __CD exists:Content-Disposition
header __CT exists:Content-Type
header __CTE exists:Content-Transfer-Encoding
header __MIME_VERSION exists:MIME-Version
header __CT_TEXT_PLAIN Content-Type =~ /^text\/plain\b/i
meta MIME_HEADER_CTYPE_ONLY (!__CD && !__CTE && __CT && !__MIME_VERSION && !__CT_TEXT_PLAIN)
describe MIME_HEADER_CTYPE_ONLY 'Content-Type' found without required MIME headers
header WITH_LC_SMTP Received =~ /\swith\ssmtp;\s/
describe WITH_LC_SMTP Received line contains spam-sign (lowercase smtp)
header FROM_NO_LOWER From:addr !~ /[a-z]/ [if-unset: x@example.com]
describe FROM_NO_LOWER From address has no lower-case characters
header SUBJ_BUY Subject =~ /^buy/i
describe SUBJ_BUY Subject line starts with Buy or Buying
header __FROM_HAS_UNDERLINE_NUMS From =~ /_\S?(?:[a-z]+\w*?\d+|\d+\w*?[a-z]+)\w*\@/i
meta FROM_HAS_ULINE_NUMS (!REPLY_TO_ULINE_NUMS && __FROM_HAS_UNDERLINE_NUMS)
describe FROM_HAS_ULINE_NUMS From: contains an underline and numbers/letters
header NIGERIAN_SUBJECT1 Subject =~ /^(?:Re:|\[.{1,10}\])?\s*(?:(?:very )?URGENT|ATTENTION)\s*$/i
describe NIGERIAN_SUBJECT1 Subject is indicative of a Nigerian spam
header NIGERIAN_SUBJECT2 Subject =~ /^(?:Re:|\[.{1,10}\])?\s*(?:very )?urgent\s+(?:(?:and|&)\s+)?(?:confidential|assistance|business|attention|reply|response|help)\b/i
describe NIGERIAN_SUBJECT2 Subject is indicative of a Nigerian spam
# this code uses an access database (sendmail, postfix, etc.)
# Since you need to actively create an accessdb to use it, the rule is
# considered userconf and is disabled by default.
header ACCESSDB eval:check_access_database('/etc/mail/access.db')
describe ACCESSDB Message would have been caught by accessdb
tflags ACCESSDB userconf
# seems to be ratware
header RCVD_AM_PM Received =~ /; [A-Z][a-z][a-z], \d{1,2} \d{4} \d{1,2}:\d\d:\d\d [AP]M [+-]\d{4}/
describe RCVD_AM_PM Received headers forged (AM/PM)
header HEADER_COUNT_CTYPE eval:check_header_count_range('Content-Type','2','999')
describe HEADER_COUNT_CTYPE Multiple Content-Type headers found
header __USER_AGENT_MSN X-Mailer =~ /^MSN Explorer /
header NO_RDNS_DOTCOM_HELO eval:check_for_no_rdns_dotcom_helo()
describe NO_RDNS_DOTCOM_HELO Host HELO'd as a big ISP, but had no rDNS
header X_ORIG_IP_NOT_IPV4 X-Originating-IP !~ /\[?(?:\d{1,3}\.){3}\d{1,3}\]?/ [if-unset: 0.0.0.0]
describe X_ORIG_IP_NOT_IPV4 X-Originating-IP doesn't look like IPv4 address
# match the format of a legit X-Auth-Warning header, and hit on fake ones
# normal: "e4e.oac.uci.edu: foo owned process doing -bs"
# fake: "bzgrdag, upaeqehv"
header X_AUTH_WARN_FAKED X-Authentication-Warning !~ /^(\S+\.\S+): / [if-unset: host.example.net: foo owned process doing -bs]
describe X_AUTH_WARN_FAKED X-Authentication-Warning header looks faked
# host no longer exists according to administrator
header FAKE_OUTBLAZE_RCVD Received =~ /\.mr\.outblaze\.com/
describe FAKE_OUTBLAZE_RCVD Received header contains faked 'mr.outblaze.com'
# domains never longer used for email, confirmed by administrator
header FROM_NONSENDING_DOMAIN From:addr =~ /\@(?:altavista\.com|eudora\.com)$/i
describe FROM_NONSENDING_DOMAIN Message is from domain that never sends email
header SUBJ_2_NUM_PARENS Subject =~ /^\(\d+\).*\(\d+\)\s*$/
describe SUBJ_2_NUM_PARENS Subject contains common spam sign (2 numbers)