CALLBACK
TOAD
VISHING
REMOTE TOOLS
DetectionBlue TeamSOCOctober 12, 2026•9 min read

Detecting Callback Phishing and TOAD

There is nothing malicious in the email. The attack starts when the victim picks up the phone, so you detect the lure's shape and the tool that follows.

Scroll

In telephone-oriented attack delivery the email carries no link or attachment, only an invoice, subscription renewal or fraud alert with a number to call. A scammer on the line walks the victim into installing a remote-access tool. Content scanners find little, so detection leans on message shape and on endpoint telemetry for remote tools.

Signals at a Glance

SignalWhereATT&CK
Invoice, renewal or charge lure with no URL and no attachmentEmail gateway, message traceT1566
Phone number in the body from a free-mail or first-contact senderGateway body rulesT1566
Quick Assist, AnyDesk, ScreenConnect or similar started by a user without a ticketSysmon 1, EDRT1219
Remote tool installer or portable binary fetched shortly before executionProxy, EDRT1105
External chat or call from an unknown tenant posing as help deskCollaboration audit logsT1566

Message Shape

Without a payload, hunt on the combination of urgency subject, no links, no attachments and an external sender the recipient has not seen before. Treat the output as a hunt list to enrich, not an alert on its own:

// Defender XDR (KQL) - payload-free external mail with a billing lure
let withUrl = EmailUrlInfo | where Timestamp > ago(7d) | distinct NetworkMessageId;
let withAtt = EmailAttachmentInfo | where Timestamp > ago(7d) | distinct NetworkMessageId;
EmailEvents
| where Timestamp > ago(7d)
| where EmailDirection == "Inbound" and DeliveryAction == "Delivered"
| where Subject matches regex @"(?i)(invoice|receipt|renewal|subscription|payment|charge|order|refund|fraud)"
| join kind=leftanti withUrl on NetworkMessageId
| join kind=leftanti withAtt on NetworkMessageId
| summarize recipients=dcount(RecipientEmailAddress), subjects=make_set(Subject, 3)
          by SenderFromDomain, SenderIPv4
| where recipients >= 3
| order by recipients desc
// tune/validate: transactional mail from real vendors lacks links too; allowlist known billers

The Remote Tool Follows the Call

The compromise begins when the caller guides the victim through installing or enabling a remote tool. Build a list of sanctioned remote-access software and alert on every other tool, especially on user workstations in the hours after a lure hits:

// Defender XDR (KQL) - remote-access tools started on endpoints outside the approved set
let approved = dynamic(["teamviewer.exe"]);    // replace with your sanctioned tools
DeviceProcessEvents
| where Timestamp > ago(7d)
| where FileName in~ ("quickassist.exe","anydesk.exe","screenconnect.clientservice.exe","screenconnect.windowsclient.exe",
                      "atera agent.exe","splashtopstreamer.exe","rustdesk.exe","remoteutilities.exe","teamviewer.exe")
| where FileName !in~ (approved)
| summarize firstSeen=min(Timestamp), launches=count(), parents=make_set(InitiatingProcessFileName, 3)
          by DeviceName, AccountName, FileName
| order by firstSeen desc
// tune/validate: Quick Assist ships with Windows; consider blocking it for users with no help-desk role
Vetted sources

Use SigmaHQ remote-tool rules, Elastic's remote-access software detections, CISA guidance on remote-monitoring tools abuse and your email gateway's message-trace guidance. Baseline your own environment — remote support tools are used legitimately by IT and vendors.

Build Detections That Fire

Mjolnir Security helps blue teams turn threat intelligence into high-fidelity detections across endpoint, network, identity and cloud telemetry.

Detection EngineeringManaged Detection & ResponseThreat HuntingIncident ResponsePurple Teaming
  • Detection Engineering We build and tune detections mapped to ATT&CK from your own telemetry.
  • Threat Hunting as a Service Hypothesis-driven hunts for the behaviours on this page.
  • 24/7 Incident Response Containment when a detection fires. Call +1 833 403 5875.
Written by: Mjolnir Threat Research (MTAC)  |  Published: October 12, 2026  |  Last reviewed: October 12, 2026