Skip to navigation

Threat reportRansomwareTL-2026-6390

Ransomware initial access brokers shift deals from public forums to private channels (Tox) as ransomware volume hits record highs

highACTIVE

ZeroFox data cited by Help Net Security shows ransomware/extortion gangs claiming about 718 victims per month through October 2026 (vs about 573/month in 2025), with at least 2,381 incidents in Q3 2026. After law enforcement took down XSS (July 2025), RAMP (January 2026) and LeakBase (March 2026), public initial-access listings fell from about 620 to about 370 per quarter as brokers moved sales to private channels, most commonly Tox.

Key facts

IdentifiersTL-2026-6390
SeverityHIGH
Affected productsSalesforce Salesforce CRM (via third-party connected apps: Salesloft Drift, Gainsight, Klue); Microsoft Microsoft Teams (external-contact vishing)
MalwareAgendaCrypt, Icarus, the gentlemen
ToolsKlue Battlecards Salesforce connected app
MITRE ATT&CK techniquesT1078 Valid Accounts, T1199 Trusted Relationship, T1213 Data from Information Repositories, T1528 Steal Application Access Token, T1539 Steal Web Session Cookie, T1550.001 Use Alternate Authentication Material: Application Access Token, T1566.004 Phishing: Spearphishing Voice, T1567 Exfiltration Over Web Service, T1650 Acquire Access, T1657 Financial Theft
MotivationFINANCIAL
Target sectorstechnology, finance, health, manufacturing, government administration
Target regionsNorth America, Asia-Pacific, 005 - South America, Europe
Detection rules9 (1 shown free)
Sources8 cited sources

Detection coverage

As of 2026-11-18, Threadlinqs Intelligence publishes 9 detection rule(s) for TL-2026-6390 across Splunk SPL, Microsoft KQL and Sigma, covering 10 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: Microsoft Teams Voice Phishing - Burst of External-Tenant Chats to One User (TL-2026-6390)

Detects a user involved in three or more distinct Microsoft Teams chat threads with users from a foreign tenant within one hour. TL-2026-6390 reports Microsoft Teams voice phishing (vishing, T1566.004) reaching nearly 10x its mid-2025 weekly level; attackers impersonating IT/help desk from external tenants open chats and calls to obtain credentials, MFA approvals or remote-support sessions (T1078 follow-on). Behaviour-first: no reliance on known-bad tenant names.

  • Languages: Splunk SPL, Microsoft KQL, Sigma
  • ATT&CK: T1566.004, T1078
  • Data sources: o365:management:activity (Teams audit), Microsoft 365 Unified Audit Log - MicrosoftTeams workload
  • Severity: medium
  • Confidence: medium

Splunk SPL

index=o365 sourcetype="o365:management:activity" Workload=MicrosoftTeams Operation IN ("ChatCreated","MessageSent","MemberAdded") earliest=-24h
| where lower('ParticipantInfo.HasForeignTenantUsers')="true"
| bin _time span=1h
| stats dc(ChatThreadId) as external_chats values(Operation) as operations values(ClientIP) as client_ips by _time UserId
| where external_chats>=3
| table _time UserId external_chats operations client_ips

Microsoft KQL

CloudAppEvents
| where Timestamp > ago(24h)
| where Application == "Microsoft Teams"
| where ActionType in ("ChatCreated", "MessageSent", "MemberAdded")
| extend Raw = parse_json(RawEventData)
| where tostring(Raw.ParticipantInfo.HasForeignTenantUsers) =~ "true"
| summarize ExternalChats = dcount(tostring(Raw.ChatThreadId)), Operations = make_set(ActionType), SourceIPs = make_set(IPAddress) by AccountObjectId, AccountDisplayName, bin(Timestamp, 1h)
| where ExternalChats >= 3
| project Timestamp, AccountObjectId, AccountDisplayName, ExternalChats, Operations, SourceIPs

Sigma

title: Microsoft Teams Chat With Foreign Tenant Users (Vishing Lead)
id: 3f0b6c1e-5a52-4d1c-9b7e-1a2c8f6d4e01
status: experimental
description: Detects Teams chat creation or messaging that involves users from a foreign tenant. Use as a lead for Teams voice-phishing (help-desk impersonation) hunting; pair with SIEM-side volume thresholds. TL-2026-6390.
references:
    - https://www.helpnetsecurity.com/2026/11/19/ransomware-access-brokers-cybercrime/
author: Threadlinqs Team
date: 2026-11-18
tags:
    - attack.initial-access
    - attack.t1566.004
    - attack.t1078
logsource:
    product: m365
    service: audit
detection:
    selection:
        Workload: MicrosoftTeams
        Operation:
            - ChatCreated
            - MessageSent
            - MemberAdded
        ParticipantInfo.HasForeignTenantUsers: 'true'
    condition: selection
falsepositives:
    - Legitimate B2B collaboration with partners, vendors or customers
    - Federated mergers/acquisitions tenants
level: medium

Known false positives:

  • Legitimate B2B collaboration with partner or vendor tenants generating several external chats per hour
  • Customer-support or sales staff who routinely chat with external tenants

Tuning notes: Restrict to users who do not normally chat externally; allow-list known partner tenant IDs; tighten threshold (>=2) for help-desk and finance staff; correlate with Quick Assist/remote-support launches and password or MFA resets on the same user within 24h.

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)
  • Entra ID OAuth Consent Granting Broad Data-Access or Offline Scopes · unlock with TL Creds
  • Entra ID Session Reuse From Multiple Countries (Session Token Theft / Replay) · unlock with TL Creds
  • Dormant Service Principal or Vendor Integration Credential Suddenly Active · unlock with TL Creds
  • Salesforce Login or API Activity From Klue, Gainsight or Salesloft Drift Connected Apps · unlock with TL Creds
  • Rclone Bulk Data Transfer to Cloud Storage (Double-Extortion Exfiltration) · unlock with TL Creds
  • Ransomware Recovery Inhibition - Shadow Copy and Backup Catalog Deletion · unlock with TL Creds
  • Help-Desk Password Reset Followed by Security-Info (MFA) Registration Within 60 Minutes · unlock with TL Creds
  • Salesforce Bulk Data Extraction Through API or Report Export (OAuth Connected App Abuse) · unlock with TL Creds

Indicators of compromise

As of 2026-11-18, Threadlinqs Intelligence tracks 10 indicators of compromise for TL-2026-6390.

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

Technical analysis

This is an ecosystem-level trend report, not a single campaign. It rests on one secondary source (Help Net Security, 2026-11-19, citing ZeroFox figures); the underlying ZeroFox report was not located or independently verified.

Volume: ZeroFox counts ransomware and extortion gangs claiming about 718 victims per month through October 2026, up from about 573 per month in 2025. Q3 2026 had at least 2,381 incidents, the highest quarter on record and about 14% above the previous high of 2,091 (late 2025). October 2026 alone had 852 incidents, and the 2026 total passed all of 2025 by the end of October. Qilin was the most active group (at least 1,104 of 7,239 incidents logged through 2026-11-02, about 15%), followed by The Gentlemen (935). North America remained the most targeted region, but its share fell from about 59% (H2 2025) to 46-48% per month since July 2026, while Asia-Pacific and South America reached their highest shares since at least early 2024.

Initial-access market: public dark-web listings selling access to compromised corporate networks fell from about 620 per quarter (early 2025) to about 370 (early 2026). Researchers say this very likely does not reflect less crime. Over the same period law enforcement took down XSS (July 2025; the suspected administrator was arrested in Kyiv on 2025-07-22), RAMP (seized by the FBI on 2026-01-28) and LeakBase (Europol-coordinated action on 2026-03-03/04). Brokers moved deals into private channels, with the encrypted messenger Tox the most common choice, and some ransomware groups appear to be handling initial access themselves. Basic access averaged about $1,427 in early 2023 and sold for under $500 by Q3 2026; domain administrator credentials averaged about $8,000. Researchers expect more session-token listings (which bypass sign-in, including MFA) and premium pricing for access to vendors and integration providers, since one compromise can reach many customers.

Related access vectors named by the source: (1) OAuth token compromises attributed to actors using the ShinyHunters name between August 2025 and June 2026 at Salesloft Drift, Gainsight and Klue, which gave access to customers' Salesforce data through authorized sessions that sign-in monitoring did not flag; Huntress attributed the June 2026 Klue intrusion to an extortion group calling itself Icarus, based on matching Session Messenger IDs; (2) malicious voice phishing over Microsoft Teams, which by mid-2026 reached nearly 10 times its mid-2025 weekly level. The source gives no CVEs, malware families, network IOCs or technical TTP detail. The IOC list therefore contains only named actors, platforms and behaviors, and no BeaconBeagle correlation was possible. Defender takeaways from the source: inventory which vendors hold standing tokens into SaaS tools, detect anomalous sessions, harden help-desk identity verification for password resets, and treat a leak-site listing as a lead that needs confirming, since unverified or exaggerated claims are expected to become routine in 2027.

MITRE ATT&CK techniques by tactic:

Initial Access

T1078 Valid Accounts; T1199 Trusted Relationship; T1566.004 Phishing: Spearphishing Voice

Collection

T1213 Data from Information Repositories

Credential Access

T1528 Steal Application Access Token; T1539 Steal Web Session Cookie

lateral-movement

T1550.001 Use Alternate Authentication Material: Application Access Token

Exfiltration

T1567 Exfiltration Over Web Service

Resource Development

T1650 Acquire Access

Impact

T1657 Financial Theft

Remediation

Immediate actions

  • Inventory third-party and SaaS integrations holding standing OAuth tokens (for example into Salesforce) and revoke tokens for unused or suspicious apps
  • Review SaaS audit logs for sessions from connected apps that behave abnormally, and rotate credentials where integration activity looks anomalous
  • Require out-of-band identity verification at the help desk before password or MFA resets
  • Restrict external Microsoft Teams communication and warn staff about vishing over Teams

Workarounds

  • Limit scopes and lifetimes of OAuth connected apps and require admin approval for new integrations

Longer-term hardening

  • Detect session-token theft and replay with token-binding and conditional-access policies
  • Monitor for compromised credentials and access offers tied to your organization
  • Treat leak-site claims as leads and validate them before declaring an incident
  • Include vendor and integration-provider compromise in third-party risk and incident-response planning

Affected and fixed versions

  • Salesforce — Salesforce CRM (via third-party connected apps: Salesloft Drift, Gainsight, Klue)
    Vulnerable versions: Not a platform vulnerability; customers with compromised OAuth integrations
  • Microsoft — Microsoft Teams (external-contact vishing)
    Vulnerable versions: Not a product vulnerability; social-engineering target

Timeline

  • Suspected administrator of the Russian-language XSS forum arrested in Kyiv by Ukrainian, French and Europol authorities; the forum domain was seized.
  • Per the source, ShinyHunters-branded OAuth token compromises (Salesloft Drift, then Gainsight, then Klue) begin in August 2025 and run through June 2026 (exact start day not stated; first of month used as placeholder).
  • Salesforce revoked all access and refresh tokens tied to Gainsight-published apps after unusual activity; Google Threat Intelligence linked it to ShinyHunters-tied actors.
  • FBI seized the RAMP ransomware-friendly forum; domains now point to FBI seized nameservers.
  • Europol-coordinated action against LeakBase (14 countries); about 100 enforcement actions against 37 active users on March 3, domain seized on March 4.
  • Attackers first accessed Klue's backend via a long-dormant API credential and pushed malicious code to harvest customers' OAuth tokens.
  • Huntress published its account of the Klue breach, attributing activity to the extortion group Icarus via matching Session Messenger IDs.
  • Q3 2026 closes with at least 2,381 ransomware/extortion incidents, the highest quarter on record and about 14% above the prior high of 2,091.
  • October 2026 recorded 852 incidents; the 2026 total passed all of 2025 by the end of October.
  • Help Net Security published the ZeroFox-based analysis of access brokers moving deals off public forums to private channels such as Tox.

Sources cited

Further reading

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

Threadlinqs Intelligence — Real-Time Threat Detection Platform

// threat_feed
$ sort --newest
Showing all threats

Live intelligence console

Threat level

Latest Threats