Threat Intelligence

The Zoho Sign Request That Passed Every Check Except the Reply-To: Government Impersonation via E-Sign Infrastructure

Written by Audian Paxson | May 26, 2026 11:00:00 AM
TL;DR A digital signature request arrived from notifications@zohosign[.]com through Zoho transmail infrastructure (system-60.transmail[.]net, IP 135.84.80[.]60). SPF, DKIM, DMARC, and ARC all passed. The message used a standard Zoho Sign template with a green banner, logo, and prominent 'Review' CTA button. Every link in the message pointed to sign.zoho[.]com with a valid sign_id parameter, plus standard guest-reminder and report-abuse Zoho links. All scanned clean. The Reply-To header, however, was set to info@admin-director[.]info, an unverifiable domain with no discoverable WHOIS data and no published DMARC policy. The in-message 'Sender' field also displayed an admin-director[.]info address. The document claimed to originate from a Connecticut government office. The organization name was listed as 'Information_Department' with an underscore, inconsistent with government communications standards. The expiry date was May 20, 2026.
Severity: High Credential Harvesting Impersonation Platform Abuse MITRE: {'id': 'T1566.002', 'name': 'Phishing: Spearphishing Link'} MITRE: {'id': 'T1036.005', 'name': 'Masquerading: Match Legitimate Name or Location'} MITRE: {'id': 'T1585.001', 'name': 'Establish Accounts: Social Media Accounts'}

A Zoho Sign document review request arrived with perfect authentication. SPF passed. DKIM passed. DMARC passed. ARC validated across every hop. The green banner, the logo, the "Review and Sign" button all matched a standard Zoho Sign template. Every link in the message resolved to sign.zoho[.]com. There was nothing for a gateway to flag.

Except the Reply-To pointed to info@admin-director[.]info, a domain with no WHOIS data, no DMARC, and no verifiable connection to any government office.

Zoho Sign as the Authenticated Sender

The email originated from notifications@zohosign[.]com through system-60.transmail[.]net at IP 135.84.80[.]60. Zoho's transactional mail infrastructure handled delivery into the recipient's Microsoft 365 environment. SPF passed because Zoho's servers are authorized senders for zohosign[.]com. DKIM passed under Zoho's signing keys. DMARC aligned. ARC seals validated.

This is the same authentication profile that every legitimate Zoho Sign notification carries. A secure email gateway evaluating sender reputation, authentication results, and link destinations would find a fully trusted platform delivering a routine document signing request. The infrastructure was legitimate because Zoho Sign actually sent the email.

Every Link Clean, Every Domain Trusted

The message contained a primary "Review" CTA linking to sign.zoho[.]com with a sign_id parameter. Additional links included a guest reminder URL and a report-abuse link, both Zoho-hosted. All resolved to Zoho infrastructure. All scanned clean.

There were no external redirects, no newly registered domains in the link chain, and no attachments. A content scanning gateway inspecting URLs and payloads would find nothing actionable. The attack surface was not in the links.

The Reply-To That Exposed the Impersonation

The From header showed notifications@zohosign[.]com. The Reply-To header showed info@admin-director[.]info. The in-message "Sender" field also displayed an admin-director[.]info address. That domain has no discoverable WHOIS registration data and no published DMARC policy.

The document claimed to come from a Connecticut government office. The organization was listed as Information_Department, underscore included. Government agencies do not use underscores in their official names. This formatting detail, combined with the unverifiable domain, points to an impersonation built quickly with a platform account rather than through any legitimate government workflow.

The document carried an expiry date of May 20, 2026, creating time pressure to act before verifying. This is standard social engineering: impose a deadline, invoke authority, and deliver through a platform the recipient trusts.

What Behavioral Detection Identified

Themis, the IRONSCALES Adaptive AI engine, evaluates the gap between what the email claims and what the headers reveal. The Reply-To mismatch between a Zoho platform sender and an unverifiable external domain, the government authority claim with no supporting infrastructure, and the formatting inconsistencies in the organization name are behavioral signals that authentication cannot assess. These patterns identify impersonation at the intent layer, not the infrastructure layer.

Authentication confirmed that Zoho Sign sent this email. It did not confirm who created the signing request or whether the document was legitimate.

See Your Risk: Calculate how many threats your SEG is missing

Indicators of Compromise

TypeIndicatorContext
Sending Platformnotifications@zohosign[.]comLegitimate Zoho Sign sender
Sending Relaysystem-60.transmail[.]netZoho transactional mail infrastructure
Sending IP135.84.80[.]60Zoho transmail IP
Reply-Toinfo@admin-director[.]infoUnverifiable domain, no WHOIS, no DMARC
In-Message Senderadmin-director[.]info addressMatches Reply-To diversion target
Document Linkssign.zoho[.]com with sign_id parameterAll Zoho-hosted, scanned clean
Organization NameInformation_DepartmentUnderscore formatting, inconsistent with government communications
Claimed AuthorityConnecticut government officeNot verifiable
Document ExpiryMay 20, 2026Urgency mechanism

MITRE ATT&CK Mapping

TechniqueIDRelevance
Phishing: Spearphishing LinkT1566.002Zoho Sign document review link as delivery vector
Masquerading: Match Legitimate Name or LocationT1036.005Government office impersonation through e-sign platform
Establish Accounts: Social Media AccountsT1585.001Zoho Sign account created to send fraudulent signing requests
Email Attack of the Day is a daily series from IRONSCALES spotlighting real phishing attacks caught by Adaptive AI and our community of 30,000+ security professionals. Each post breaks down one attack — what it looked like, why it worked, and what you can do about it.