Anatomy of a Coined-Brand E-Signature Phishing Kit
Anatomy of a Coined-Brand E-Signature Phishing Kit
Since May 2025, one operator has run more than 20 coined e-signature brands as a single phishing kit whose landing-page fingerprint outlasted four mail rails. Each domain wears an invented signing-service name (bellasign[.]net, voltsign[.]net, swiftverify[.]net) with no real trademark to infringe, so brand-abuse takedowns have nothing to grab. Behind the different names sits one build: the same landing-page host pattern, the same mailing-list header, the same first-name-personalized "your signature is required" template. The operator has swapped email providers three times to stay deliverable, yet the kit underneath never changed. This is a study of that kit: how the domains are coined, how they cluster in registration records, and which signals tie every one of them back to a single hand.
Key Takeaways
- One operator has run more than 20 purpose-built domains as a single e-signature phishing kit since May 2025, using coined
*signand*verifynames that carry no real trademark for takedown teams to act on. - The durable fingerprint is an
i.<domain>/<7-character shortcode>landing host paired with alistrequest@<domain>List-Unsubscribe header, present on every domain in the network. - The operator rotated across four legitimate email providers (Mailjet, Amazon SES, Postmark, Mandrill) for deliverability and takedown resilience while the landing pages and email templates stayed constant.
- Domains were batch-registered in same-day cohorts, first through Namecheap and then Cloudflare Registrar, all behind WHOIS privacy, which is a strong single-operator signal.
- The lure evolved from a plain "sign this document" notice into invoice and funding payoffs carrying fabricated dollar amounts and transaction IDs.
Background
Fake e-signature requests are a well-worn phishing pretext. The Identity Theft Resource Center has warned consumers about emails impersonating "DocuSign Electronic Service" that carry fake notification lures and lead to credential theft, and OneSpan has documented criminals exploiting DocuSign-style envelopes against businesses and government. University IT offices such as Cornell publish standing guidance on verifying whether a signing request is real, Docusign maintains public safety alerts about brand-impersonation, and U.S. healthcare-sector guidance from HHS HC3 has issued its own alert on DocuSign abuse. The pretext is recognized. What makes this network worth a closer look is that it does not impersonate any of those brands by name. It coins its own.
The operator delivers exclusively through legitimate email service providers (ESPs), which is the first piece of context a reader needs. An ESP is a hosted platform that sends mail on a customer's behalf from warmed, reputable IP ranges with valid SPF and DKIM authentication. Scammers onboard to these platforms because a message that authenticates cleanly and originates from a widely-allowlisted sender lands in the inbox rather than the spam folder, and because self-service signup lets them create, burn, and replace accounts faster than abuse teams close them. This operator used four:
- Mailjet, a self-service transactional and marketing API and SMTP relay (Sinch-owned), reached first. Free signup gives immediate access to shared reputable IPs, and each account gets its own click-tracking subdomain on
mjt[.]luwith bounce handling onbnc3.mailjet[.]com. - Amazon SES, AWS's pay-as-you-go email service, which inherits Amazon's IP reputation and wraps links through its own
awstrack[.]metracking domain. - Postmark, a developer-focused provider known for strong inbox placement, which uses a
pm-bounces.<domain>bounce subdomain convention. - Mandrill, Mailchimp's transactional add-on riding Mailchimp reputation, with tracking served through
mandrillapp[.]com.
Every one of those tracking and bounce domains is legitimate ESP infrastructure. Its appearance in a message tells you which rail carried the mail, not who owns the sending domain. The operator registered the sending domains themselves through Cloudflare Registrar and Namecheap, both of which offer cheap bulk registration behind default WHOIS privacy, which is exactly what a throwaway-domain operation wants.
Discovery and Infrastructure
The network first surfaced as a handful of unrelated-looking signing-notification domains. Clustering them took a fingerprint rather than a shared IP or a shared registrant, because there is neither: the domains sit behind privacy proxies and route through shared ESP infrastructure. What ties them together is the build. Every domain resolves the same three subdomains (app.<domain>, i.<domain>, and the bare apex), serves its landing pages from the i. host under a seven-character shortcode path, and stamps a listrequest@<domain> address into the List-Unsubscribe header. Pivoting on that combination, plus the coined *sign / *verify naming and the first-name signing subject, grew the confirmed set to more than 20 apexes.
The domain inventory, with public WHOIS registration data:
| Domain | Registrar | Created | Brand family |
|---|---|---|---|
miarvo[.]net |
Namecheap | 2025-05-05 | (earliest, outlier) |
olesign[.]net |
Namecheap | 2025-05-06 | *sign |
signrise[.]net |
Namecheap | 2025-05-06 | *sign |
swiftsign[.]net |
Namecheap | 2025-05-16 | *sign |
dotsign[.]net |
Namecheap | 2025-05-16 | *sign |
servicebeehive[.]com |
Cloudflare | 2025-05-21 | (generic) |
upverify[.]net |
Cloudflare | 2025-06-04 | *verify |
swiftverify[.]net |
Cloudflare | 2025-06-04 | *verify |
aloverify[.]com |
Cloudflare | 2025-06-30 | *verify |
suavesign[.]com |
Cloudflare | 2025-06-30 | *sign |
voltsign[.]net |
Cloudflare | 2025-08-13 | *sign |
sealsnap[.]net |
Cloudflare | 2025-08-13 | (seal metaphor) |
bellasign[.]net |
Cloudflare | 2025-08-19 | *sign |
sigtouch[.]com |
Cloudflare | 2025-08-19 | (sig metaphor) |
unoverify[.]com |
Cloudflare | 2025-08-19 | *verify |
monosign[.]net |
Cloudflare | 2025-09-19 | *sign |
venusign[.]com |
Cloudflare | 2025-11-03 | *sign |
splashsign[.]com |
Cloudflare | 2025-11-08 | *sign |
verasign[.]net |
Cloudflare | 2025-12-08 | *sign / *verify |
| ... | verified-malicious subset; the fingerprinted network spans more than 20 operator apexes |
Each of these carries a support@ sending address and an i.<domain> landing host. None resolves to a real business. Where a landing page still answers, it renders a thin "Modern E-Signatures for Modern Teams" facade with no contact details, no pricing, no login, and no legal pages, which is the shell, not a product.
How It Works
A representative contact chain runs like this. The target receives a short, personalized email from a support@ or notification@ address on one of the coined domains, delivered through whichever ESP the operator is using that week. The subject carries the recipient's first name and a signing cue: "Review and Sign Document [name]," "Signing notification [name]," "Record ready for digital signature, [name]." The body is a few vague sentences referencing an "application," an "approval," or a "document in the system," ending with a single call-to-action button.
That button points at i.<domain>/<7-character shortcode>, either directly or wrapped through the ESP's click-tracker. The shortcode is unique per recipient, which lets the operator tie a click back to a specific send. The landing host presents the fake signing portal that harvests whatever the victim enters to "sign." The footer offers a listrequest@<domain> unsubscribe and, in many variants, a plain-language opt-out: "reply with 'No longer needed'." That opt-out doubles as an engagement probe, since a reply confirms a live, attentive mailbox.
Sample Lures
Three template families run across the network. All recipient-identifying data has been replaced with placeholders; the fabricated dollar amounts, transaction IDs, and signer names below are attacker-authored lure content, not victim data.
Template A, the standard signing notice (defanged, delivered via Mailjet):
From: "Support" <support@sealsnap[.]net>
Subject: Review and Sign Document [recipient first name]
Reply-To: support@sealsnap[.]net
List-Unsubscribe: <mailto:listrequest@sealsnap[.]net>
Return-Path: <bounce@a3238981.bnc3.mailjet[.]com>
Here is the placed status. Your Document exists in the system. To bring this
process to its conclusion your signature is required.
[ Continue to Signing ] -> http[:]//i.sealsnap[.]net/<shortcode>
Grace Morgan
Template B, the more formal "Electronic Signature Processing" variant. Its wording appears verbatim on multiple domains, which is one of the tighter attribution links in the set:
From: "Support Team" <support@bellasign[.]net>
Subject: Document revision inquiry
List-Unsubscribe: <mailto:listrequest@bellasign[.]net>
Electronic Signature Processing request. The relevant documents have been
received related to your application and our team is currently reviewing them.
[ Proceed to Document ] -> http[:]//i.bellasign[.]net/<shortcode>
Allen Austin
Template C, the evolved invoice and funding payoff, which adds a fabricated dollar amount and a transaction ID to manufacture financial urgency:
From: "Support" <support@miarvo[.]net>
Subject: Inquiry received successfully
List-Unsubscribe: <mailto:listrequest@miarvo[.]net>
Digital signature needed - $2,520.00 | [recipient name] | Apr 11, 2026 |
V282-52A - Application Verification.
[recipient first name], the inbound funding details for $2,520 reached our
intake system.
[ CONTINUE ] -> http[:]//i.miarvo[.]net/<shortcode>
Audrey
Technical Analysis
Coined-Brand Domain Generation
The naming is deliberate. Instead of typo-squatting a real signing service, which invites a fast brand-abuse takedown, the operator coins its own trust-flavored names from a small vocabulary. Two lexical families dominate. The *sign family (bellasign, voltsign, suavesign, olesign, dotsign, monosign, splashsign, swiftsign, signrise, venusign) pairs a short evocative prefix with a signing suffix. The *verify family (swiftverify, unoverify, upverify, aloverify) does the same with a verification suffix. A few sit between the two poles: sigtouch and sealsnap lean on signing and seal metaphors, verasign blends "verify" and "sign," and servicebeehive and the earliest domain miarvo[.]net fall outside the pattern entirely. Every coined name reads as a plausible small SaaS brand, none collides with an existing trademark, and each is short enough to look native inside a support@<brand> sender.
Registration Cohorts and Batch Co-Registration
Registration records are where a single operator becomes hard to argue against. The domains cluster into same-day cohorts, several registered in tight batches alongside one another:
| Cohort | Registrar | Domains created together |
|---|---|---|
| 2025-05-05 to 05-16 | Namecheap | miarvo, olesign, signrise, swiftsign, dotsign |
| 2025-06-04 | Cloudflare | upverify[.]net + swiftverify[.]net (same day) |
| 2025-06-30 | Cloudflare | aloverify[.]com + suavesign[.]com (same day) |
| 2025-08-13 | Cloudflare | voltsign[.]net + sealsnap[.]net (same day) |
| 2025-08-19 | Cloudflare | bellasign[.]net + sigtouch[.]com + unoverify[.]com (same day) |
| 2025-09 to 12 | Cloudflare | monosign, venusign, splashsign, verasign (staggered) |
The operator opened on Namecheap in early May 2025, then migrated to Cloudflare Registrar within weeks and stayed there. Same-day pairs and trios of otherwise-unrelated coined names, all behind WHOIS privacy, do not happen by coincidence across independent registrants. The staggered late-2025 additions show the operation restocking its domain inventory on a rolling basis rather than in one burst, consistent with a build that expects steady attrition from takedowns.
Subdomain Grammar and the Landing-Page Fingerprint
Every apex resolves an identical three-subdomain layout: app.<domain>, i.<domain>, and the bare apex. The i. host is the constant that matters. It serves the phishing landers under a seven-character alphanumeric shortcode path (/ppe5sj5, /z2e6sgy, /armpmj6, /tf3gtnb), one per recipient. Paired with the listrequest@<domain> List-Unsubscribe header, this two-part signature (a per-recipient i.<domain>/<shortcode> CTA plus a domain-hosted listrequest@ unsubscribe) is the single most reliable way to recognize the network, because it survives every other rotation the operator performs.
Sender Persona Rotation
Early domains sent from a single support@ mailbox. Later ones spread across multiple personas on the same apex to imitate a staffed operation. olesign[.]net alone sends as support@, details@, notification@, systemnotification@, and update@. servicebeehive[.]com uses support@, solutionsspecialist@, and ccmanager@. upverify[.]net uses assistant@, details@, and notification@. The multi-persona mailboxes cluster on the newer domains, which suggests the kit's mail configuration matured over the campaign. One domain, olesign[.]net, went further and stamped a fabricated corporate identity, "Olesign Group, Inc.," onto a real Los Angeles high-rise address, an invented tenant at a genuine building.
ESP Rotation Across Four Mail Rails
The clearest sign of an operator optimizing for survival is the mail-provider rotation. The kit opened on Mailjet, with a distinct Mailjet account per domain (each carrying its own mjt[.]lu tracking subdomain and bnc3.mailjet[.]com bounce path), then added three more rails while the lander and templates held constant:
| Mail rail | Observed infrastructure | Role in the kit |
|---|---|---|
| Mailjet | per-account mjt[.]lu tracker, bnc3.mailjet[.]com bounce |
original delivery rail |
| Amazon SES | amazonses[.]com return-path, awstrack[.]me click-wrap |
added for scale and reputation |
| Postmark | pm-bounces.<domain> bounce subdomain |
added for inbox placement |
| Mandrill | mandrillapp[.]com tracking |
added rail |
Provisioning a separate ESP account per domain, rather than blasting all domains from one account, is a deliberate blast-radius control: an abuse report against one sending domain burns one account, not the fleet. That the templates and i.<domain> landers did not change across any of these migrations is what confirms the rails are interchangeable plumbing under one unchanged kit.
Template Families and Lure Evolution
The three template families trace the operator's tuning over time. Template A is the plain signing notice. Template B reframes the same ask as a formal "Electronic Signature Processing request," and its verbatim reuse across domains (bellasign[.]net and olesign[.]net carry identical wording) is a direct attribution link. Template C is the escalation: it drops in a fabricated dollar figure and a transaction ID and reframes the signature as the last step to release "inbound funding," moving the pretext from a bureaucratic chore to a financial payoff. The progression from a document chore to a funding hook shows an operator actively testing which framing converts.
Detection Observations
The behavioral signature that separates this traffic from legitimate signing notifications is compact and stable. A message pairing a coined *sign or *verify sending domain with an i.<domain>/<7-character shortcode> call-to-action and a listrequest@<domain> List-Unsubscribe header is characteristic of the network regardless of which ESP delivered it. First-name-only personalization in the subject, vague references to an unnamed "application" or "document in the system," and multiple sender personas on a freshly-registered privacy-proxied domain reinforce the read.
Because the operator rotates mail providers freely, the ESP artifact (a mjt[.]lu, awstrack[.]me, pm-bounces, or mandrillapp[.]com element) is not itself a useful selector: those are legitimate services that carry vast amounts of real mail. The cross-rail constant, the i.<domain> lander fingerprint, is the strongest pivot for defenders, since it is the one attribute the operator has never changed. Notably, the coined domains carried no reputation in public threat-intelligence feeds during their active window, which reflects how new, purpose-built, low-volume infrastructure stays quiet on external sources and argues for behavioral clustering over feed lookups.
MITRE Fight Fraud Framework Mapping
The mapping below aligns to the MITRE Fight Fraud Framework (F3) (https://ctid.mitre.org/fraud). The matrix's live technique identifiers could not be independently confirmed at time of writing, so techniques are given by tactic and behavior rather than by a possibly-incorrect ID.
| Tactic | Technique / behavior |
|---|---|
| Resource Development | Acquire purpose-built domains through registrars with WHOIS privacy |
| Resource Development | Establish multiple ESP sending accounts for rotation and blast-radius control |
| Initial Access | Deliver personalized phishing email through legitimate mail providers |
| Initial Access | Impersonate the e-signature service category with coined brand names |
| Initial Access | Add first-name personalization, fabricated transaction IDs, and dollar amounts as legitimacy cues |
| Execution | Apply signing-urgency pressure ("your signature is required") |
| Monetization | Harvest credentials through a fake document-signing portal |
Indicators of Compromise
All indicators are defanged. Recipient and victim data has been removed. Shared ESP tracking and bounce subdomains are intentionally omitted, since they belong to legitimate mail providers rather than the operator.
Domains
| Value | Role | Notes |
|---|---|---|
miarvo[.]net |
Sending / landing | Namecheap, created 2025-05-05 (earliest) |
olesign[.]net |
Sending / landing | Namecheap, 2025-05-06; verbatim Template B; fabricated "Olesign Group, Inc." |
signrise[.]net |
Sending / landing | Namecheap, 2025-05-06 |
swiftsign[.]net |
Sending / landing | Namecheap, 2025-05-16 |
dotsign[.]net |
Sending / landing | Namecheap, 2025-05-16 |
servicebeehive[.]com |
Sending / landing | Cloudflare, 2025-05-21; multi-persona senders |
upverify[.]net |
Sending / landing | Cloudflare, 2025-06-04 |
swiftverify[.]net |
Sending / landing | Cloudflare, 2025-06-04 (same day as upverify) |
aloverify[.]com |
Sending / landing | Cloudflare, 2025-06-30 |
suavesign[.]com |
Sending / landing | Cloudflare, 2025-06-30 (same day as aloverify) |
voltsign[.]net |
Sending / landing | Cloudflare, 2025-08-13 |
sealsnap[.]net |
Sending / landing | Cloudflare, 2025-08-13 (same day as voltsign) |
bellasign[.]net |
Sending / landing | Cloudflare, 2025-08-19; verbatim Template B |
sigtouch[.]com |
Sending / landing | Cloudflare, 2025-08-19; Template C funding payoff |
unoverify[.]com |
Sending / landing | Cloudflare, 2025-08-19 |
monosign[.]net |
Sending / landing | Cloudflare, 2025-09-19 (distinct from MonoFor MonoSign) |
venusign[.]com |
Sending / landing | Cloudflare, 2025-11-03 |
splashsign[.]com |
Sending / landing | Cloudflare, 2025-11-08 |
verasign[.]net |
Sending / landing | Cloudflare, 2025-12-08 (distinct from VeriSign Inc.) |
| ... | (verified-malicious subset; the fingerprinted network spans more than 20 operator apexes) |
Landing Hosts
| Value | Role | Notes |
|---|---|---|
i.sealsnap[.]net |
Landing | Credential-harvest portal, /<shortcode> path |
i.bellasign[.]net |
Landing | Credential-harvest portal |
i.miarvo[.]net |
Landing | Credential-harvest portal |
i.swiftverify[.]net |
Landing | Credential-harvest portal |
i.unoverify[.]com |
Landing | Credential-harvest portal |
i.suavesign[.]com |
Landing | Credential-harvest portal |
i.voltsign[.]net |
Landing | Credential-harvest portal |
i.olesign[.]net |
Landing | Credential-harvest portal |
i.sigtouch[.]com |
Landing | Credential-harvest portal |
i.verasign[.]net |
Landing | Credential-harvest portal |
| ... | (representative subset; dozens of verified-malicious i.<domain> landing hosts across the network) |
Senders
| Value | Role | Notes |
|---|---|---|
support@sealsnap[.]net |
Sender | Template A |
support@bellasign[.]net |
Sender | Template B |
support@miarvo[.]net |
Sender | Template C funding payoff |
support@swiftverify[.]net |
Sender | *verify family |
support@unoverify[.]com |
Sender | *verify family |
support@sigtouch[.]com |
Sender | Template C funding payoff |
details@olesign[.]net |
Sender | Multi-persona domain |
notification@olesign[.]net |
Sender | Multi-persona domain |
systemnotification@olesign[.]net |
Sender | Multi-persona domain |
assistant@upverify[.]net |
Sender | Multi-persona domain |
ccmanager@servicebeehive[.]com |
Sender | Multi-persona domain |
solutionsspecialist@servicebeehive[.]com |
Sender | Multi-persona domain |
| ... | (representative subset; 50+ verified-malicious sender addresses across the network) |
Conclusion
The lesson of this kit is that a name is disposable but a build is not. By coining its own signing brands, the operator sidesteps the trademark-based takedowns that clip typosquatters, and by swapping ESPs it keeps mail flowing as individual accounts burn. What it cannot easily change is the machine underneath: the i.<domain>/<shortcode> lander, the listrequest@ header, the personalized signing subject. Defenders watching for that fingerprint, rather than for any one domain or mail provider, will catch the next cohort the day it registers.
Related research
The Supplement SMS Kit That Borrows Real Business Identities
Since April 2026, one operation has run 26 supplement storefront domains by text message, most carrying the legal identity of a real registered business.
How a Government-Impersonation Smishing Kit Signs Its Own Registration Batches
A smishing kit impersonating toll agencies, tax offices and national police stamped the registration date into its own WHOIS records.
One Lure, Four Operators: Ad Tracking Strings as Fingerprints
Four separate operators ran the same fake TradingView desktop-app ads on Facebook for nine months, and one query-string key was all that told them apart.