Skip to navigation

Threat reportVulnerabilityTL-2026-2891

Apple iCloud Mail Parser Flaws Let Free Accounts Spoof Any @icloud.com Sender and Pass SPF/DKIM/DMARC

mediumPATCHED

Two email spoofing flaws in Apple's iCloud SMTP submission pipeline, found by Timo Longin of SEC Consult Vulnerability Lab, let any holder of a free iCloud account send mail that displayed as any @icloud.com address while passing SPF, DKIM and DMARC.

Key facts

IdentifiersTL-2026-2891
Also known asiCloud Mail From-header spoofing, From: anyone@icloud.com
SeverityMEDIUM
Affected productsApple iCloud Mail (SMTP submission infrastructure, smtp.mail.me.com)
Weaknesses (CWE)CWE-444, CWE-290
Toolssmtpsmuggling.com test tooling
MITRE ATT&CK techniquesT1566 Phishing, T1585.002 Email Accounts, T1598 Phishing for Information, T1684.001 Impersonation, T1684.002 Email Spoofing
MotivationUNKNOWN
Target sectorstechnology, finance, government administration, consumer
Target regionsGlobal
Detection rules9 (1 shown free)
Sources8 cited sources

Detection coverage

As of 2026-10-04, Threadlinqs Intelligence publishes 9 detection rule(s) for TL-2026-2891 across Splunk SPL, Microsoft KQL and Sigma, covering 8 indicator(s) of compromise. The whole corpus is readable without an account. Every threat comes with one complete detection in Splunk SPL, Microsoft KQL and Sigma, free for everyone; a free account adds daily TL Creds to unlock the rest. Paid plans add a larger daily allowance, correlation and the MCP server. Threadlinqs MCP server · View plans.

Free detection: iCloud Authenticated Sender Spoof - Return-Path/From Mismatch with SPF/DKIM/DMARC Pass

Exploitation outcome of the iCloud SMTP parser-differential flaws (CR header injection and dot-stuffing). Mail relayed by Apple's own MTA space (17.0.0.0/8) carries an @icloud.com Return-Path (the attacker's real free account) but a different @icloud.com identity in the visible From header, while SPF, DKIM and DMARC all pass. This is the primary hunting pivot named in the research. Also usable as a retro-hunt over mail received since 2024-05-21.

  • Languages: Splunk SPL, Microsoft KQL, Sigma
  • ATT&CK: T1672, T1656, T1036
  • Data sources: email_gateway_message_log (Proofpoint pps_messagelog), Microsoft Defender XDR EmailEvents, SMTP headers (Return-Path, Authentication-Results)
  • Severity: high
  • Confidence: high

Splunk SPL

``` iCloud From-header spoof: envelope identity != header identity, all auth passes ```
index=email sourcetype=pps_messagelog earliest=-24h
| eval return_path=lower(coalesce('envelope.from', return_path)), src_ip=coalesce('connection.ip', src_ip), header_from=lower(coalesce('msg.header.from{}', header_from))
| rex field=header_from "(?<hf_addr>[\w.+-]+@icloud\.com)"
| where isnotnull(hf_addr) AND like(return_path, "%@icloud.com") AND hf_addr!=return_path AND cidrmatch("17.0.0.0/8", src_ip)
| where lower(coalesce(spf,""))="pass" AND lower(coalesce(dkim,""))="pass" AND lower(coalesce(dmarc,""))="pass"
| table _time src_ip return_path hf_addr recipient subject spf dkim dmarc

Microsoft KQL

EmailEvents
| where Timestamp > ago(24h) and EmailDirection == "Inbound"
| where SenderFromDomain =~ "icloud.com" and SenderMailFromDomain =~ "icloud.com"
| where SenderFromAddress !~ SenderMailFromAddress
| where SenderIPv4 startswith "17."
| extend Auth = parse_json(AuthenticationDetails)
| where tostring(Auth.SPF) =~ "pass" and tostring(Auth.DKIM) =~ "pass" and tostring(Auth.DMARC) =~ "pass"
| project Timestamp, NetworkMessageId, SenderIPv4, SenderMailFromAddress, SenderFromAddress, RecipientEmailAddress, Subject, DeliveryAction, DeliveryLocation

Sigma

title: iCloud Sender Spoofing - Role Identity in From with Mailbox Return-Path via Apple MTA
id: 3b8f5a1e-6c4d-4f0a-9e27-1d5c7a90b2e4
status: experimental
description: Detects authenticated mail from Apple outbound infrastructure whose visible From is a role-style @icloud.com identity (no-reply, support, security, billing, admin) while the Return-Path is an individual @icloud.com mailbox. Sigma cannot compare two fields, so this approximates the exact From/Return-Path mismatch implemented in the SPL and KQL flavors.
references:
  - https://sec-consult.com/blog/detail/from-anyoneicloudcom-spoofing-arbitrary-apple-icloud-identities/
  - https://attack.mitre.org/techniques/T1672
author: Threadlinqs Team
date: 2026/10/04
tags:
  - attack.stealth
  - attack.t1672
  - attack.t1656
  - attack.t1036
logsource:
  category: email
  product: email_gateway
detection:
  selection_role_from:
    header_from|re: '(?i)\b(no-?reply|support|security|apple-?id|billing|admin|postmaster|abuse)@icloud\.com'
  selection_infra:
    return_path|endswith: '@icloud.com'
    src_ip|cidr: '17.0.0.0/8'
  selection_auth:
    spf: 'pass'
    dkim: 'pass'
    dmarc: 'pass'
  filter_self:
    return_path|re: '(?i)^(no-?reply|support|security|apple-?id|billing|admin|postmaster|abuse)@icloud\.com$'
  condition: selection_role_from and selection_infra and selection_auth and not filter_self
falsepositives:
  - iCloud users with an alias or custom role-named address that sends from a different primary mailbox
  - Mail forwarding or list software rewriting the envelope sender
level: high

Known false positives:

  • iCloud users who legitimately own multiple @icloud.com aliases and send as an alias from their primary mailbox
  • Mailing-list or forwarding software that rewrites the envelope sender while keeping the original From

Tuning notes: Map header_from, return_path, src_ip, spf, dkim and dmarc to your gateway's field names. Allow-list known alias pairs per user and known forwarders. Extend the Apple source range beyond 17.0.0.0/8 only if your gateway shows other Apple egress. The observed spoof source in the research was 17.57.155.19; it is legitimate Apple infrastructure, so never block it.

8 more detection rules (Splunk SPL, Microsoft KQL, Sigma) · unlock with TL Creds. Sign up free to unlock the detection rules or query them with the Threadlinqs MCP server (Purple)
  • High-Value iCloud Identity Impersonation (tim.cook@ / no-reply@ / admin@) with Authentication Pass · unlock with TL Creds
  • Spoofed iCloud Sender Delivering External URLs (Credential-Harvest / Phishing Lure) · unlock with TL Creds
  • SMTP Submission From-Address Rejection Probing (5.7.0 / Sender Not Owned by User) · unlock with TL Creds
  • SMTP Smuggling / Bare Line-Ending Anomalies at the Mail Transfer Agent · unlock with TL Creds
  • Header Lines Inside multipart/alternative Body and Dot-Stuffing Desync Sequence '.:' · unlock with TL Creds
  • Spoofed iCloud Identity Used for Payment-Fraud / BEC Lure Wording · unlock with TL Creds
  • Multiple or CR-Obfuscated From Headers in Inbound Mail (Parser-Differential Header Injection) · unlock with TL Creds
  • Single iCloud Account Cycling Multiple Spoofed @[indicator] From Identities · unlock with TL Creds

Indicators of compromise

As of 2026-10-04, Threadlinqs Intelligence tracks 8 indicators of compromise for TL-2026-2891.

Sample indicators (3 of 8)

  • IP address: 17.57.155[.]19
  • Domain: smtp[.]mail[.]me[.]com
  • Domain: icloud[.]com

Defanged, TLP:CLEAR. Full set of 8 indicators: unlock with TL Creds.

8 indicators of compromise · unlock with TL Creds. Compare plans

Technical analysis

The root cause was parser inconsistency (bare carriage returns in the From: header, then SMTP dot-stuffing handling). Apple has fixed both; no CVE was assigned and no in-the-wild exploitation is reported.

SEC Consult Vulnerability Lab (Timo Longin, known for the 2023 SMTP smuggling research) disclosed two vulnerabilities in Apple iCloud Mail's outbound SMTP infrastructure on 2026-10-01 (blog post 'From: anyone@icloud.com - Spoofing Arbitrary Apple iCloud Identities'; press coverage 2026-10-02). An authenticated user of the submission server smtp.mail.me.com (port 587) normally cannot send with a From: header that differs from their own address; Apple enforces this with a proprietary From-header check that rejects such submissions with '5.7.0 From address is not one of your addresses'. Both flaws bypass that check.

Flaw 1 (carriage-return line-break injection): two internal parsers handle the message differently. Parser 1 (Apple's validation layer) ignores a malformed From header that contains bare CR characters adjacent to the colon (conceptually 'From\r:\radmin@icloud.com') and so validates only the attacker's own legitimate second From header. Parser 2 (suspected Postfix based on message signatures) strips the CRs and normalizes the first header into a valid From header, so the relayed message carries a From header naming the spoofed identity. A multipart/alternative boundary was used to keep the legitimate address from being displayed. Apple's first remediation attempt reportedly relied on substring blacklisting (e.g., of 'admin') and was insufficient.

Flaw 2 (SMTP dot-stuffing inconsistency): after the first fix, the researcher found that Parser 1 does not honor the RFC 5321 section 4.5.2 dot-stuffing removal rules while Parser 2 deletes leading periods. Sequences such as '.:' therefore create an interpretation divergence that again lets a malicious From header pass validation and appear differently after relay.

Authentication still passes because the envelope sender / Return-Path remains the attacker's real iCloud address (SPF passes against Apple's own sending infrastructure, observed from 17.57.155.19), and the DKIM signature is applied after Parser 2 normalization, over the already-spoofed message; DMARC passes through alignment with icloud.com. Recipients therefore see authenticated mail apparently from identities such as tim.cook@icloud.com or no-reply@icloud.com.

Impact is trust abuse: highly credible phishing, payment-fraud/BEC and credential-harvesting mail impersonating Apple or any iCloud user that traverses gateways relying on SPF/DKIM/DMARC. Reported to Apple 2024-05-21; fixes fully rolled out and confirmed 2025-11-12. Apple paid a $15,000 bounty. No CVE or public Apple advisory was identified, no threat actor is named, and no exploitation in the wild is stated. Sourcing note: the hunt summary and the news article say fixes were confirmed in December 2025; the primary SEC Consult report gives 2025-11-12 for full rollout, which is used here. Severity (MEDIUM) and the absence of a CVSS score are analyst estimates.

MITRE ATT&CK techniques by tactic:

Initial Access

T1566 Phishing

Resource Development

T1585.002 Email Accounts

Reconnaissance

T1598 Phishing for Information

Stealth

T1684.001 Impersonation; T1684.002 Email Spoofing

Remediation

Patches

  • Server-side Apple fix; no CVE and no customer-installable patch

Immediate actions

  • No customer patch action needed for iCloud; Apple fixed the service side (confirmed fully rolled out 2025-11-12)
  • Hunt historical mail from @icloud.com for Return-Path or Authentication-Results identity that differs from the displayed From address
  • Flag messages with multiple From headers, bare CR characters in headers, or unexpected multipart/alternative boundaries containing headers

Workarounds

  • Compare Return-Path with the visible From address when triaging suspicious iCloud-origin mail

Longer-term hardening

  • Do not treat SPF/DKIM/DMARC pass as proof of sender identity for sensitive requests (credentials, payments); require out-of-band verification
  • Gateway rule: reject or quarantine RFC 5322 violations such as multiple From headers
  • For operators of SMTP submission pipelines: validate the From header against the authenticated identity in a single canonical parse (e.g., a milter) and sign DKIM only after all parsing and normalization is complete; do not rely on smtpd_sender_login_maps alone

Affected and fixed versions

  • Apple — iCloud Mail (SMTP submission infrastructure, smtp.mail.me.com)
    Vulnerable versions: Service-side, from at least 2024-05-21 until fixes rolled out; the first fix was released 2025-05-24 but the dot-stuffing variant stayed exploitable
    Fixed in: Fully remediated and confirmed by Apple 2025-11-12

Timeline

  • SEC Consult publishes Timo Longin's SMTP smuggling research, the prior art on SMTP parser differentials that bypass SPF/DKIM/DMARC.
  • Carriage-return From-header spoofing flaw reported to Apple.
  • Apple confirms it has made changes addressing the report.
  • Apple awards a $15,000 Apple Security Bounty.
  • Second spoofing vector (SMTP dot-stuffing inconsistency) discovered, bypassing the first remediation.
  • First fix released; the dot-stuffing variant remained exploitable.
  • CERT/CC publishes VU#517845 on authenticated-user From header spoofing across mail providers (Apple status listed as unknown).
  • Apple confirms fixes fully rolled out (news coverage dates confirmation to December 2025).
  • SEC Consult publishes the technical report 'From: anyone@locked-ioc-5 - Spoofing Arbitrary Apple iCloud Identities'.
  • Cyber Security News reports the flaws publicly.

Sources cited

Previous and next in vulnerability

Drafted by the Threadlinqs research pipeline from 8 sources; published ; last updated —.

Threadlinqs Intelligence — Real-Time Threat Detection Platform

// threat_feed
$ sort --newest
Showing all threats

Live intelligence console

Threat level

Latest Threats