Skip to content

WARNING

Disclaimer: This is OSINT summarized by AI. Trust, but verify.

Scattered Spider — Recent Campaigns & Evolving TTPs (2025–2026)

Date: 2026-04-07 Severity: Critical TLP: CLEAR Aliases: UNC3944, Scatter Swine, Oktapus, Octo Tempest, Storm-0875, Muddled Libra, 0ktapus, Starfraud


Executive Summary

Scattered Spider remains one of the most active and dangerous cybercriminal collectives operating in 2025–2026. Since the high-profile MGM Resorts and Caesars Entertainment breaches in late 2023, the group has significantly evolved its tactics, expanding from hospitality and telecommunications into retail, aviation, financial services, and technology sectors. Key developments include a pivot to DragonForce Ransomware-as-a-Service (RaaS), escalated Attacker-in-the-Middle (AiTM) phishing operations using Evilginx, deployment of an updated Spectre RAT with enhanced C2 capabilities, and increasingly sophisticated social engineering that targets third-party IT help desks. Despite multiple law enforcement operations resulting in arrests across the US and UK, the group's decentralized structure has allowed operations to continue with minimal disruption.

The April–May 2025 UK retail attacks (Marks & Spencer, Co-op, Harrods) caused an estimated £300M+ in damages, followed by a June–July 2025 wave targeting airlines (Hawaiian Airlines, WestJet, Qantas). A July 29, 2025 joint CISA/FBI advisory updated TTPs with new indicators including DragonForce deployment, AnyDesk/Teleport abuse, and RattyRAT usage. International law enforcement arrested multiple members in 2025, but the threat persists.


Threat Actor Profile

Structure & Organization

Scattered Spider is a loosely organized, English-speaking cybercriminal collective primarily composed of young adults (ages 17–22) from the US and UK. The group operates with a decentralized structure, recruiting members through online communities (Telegram, Discord). Members specialize in social engineering, with some actors hired specifically for vishing operations at compensation rates of $10,000–$25,000/month plus performance bonuses.

Motivation

Financial gain through data extortion, ransomware deployment, and cryptocurrency theft. The group has stolen an estimated $11M+ in cryptocurrency and caused hundreds of millions in damages through ransomware operations.

Ransomware Affiliations

  • ALPHV/BlackCat (2023): Used in MGM/Caesars attacks
  • RansomHub (2024): Affiliate operations
  • Qilin (2024): Secondary ransomware deployment
  • DragonForce (2025–present): Current primary RaaS platform; 80/20 revenue split with DragonForce operators

Campaign Timeline (2025–2026)

Q1 2025 — Phishing Infrastructure Overhaul

  • Shifted from hyphenated domains (sso-company[.]com) to subdomain-based keywords (sso[.]c0mpany[.]com) to evade domain monitoring
  • Deployed new phishing kit (#5) with dynamic subdomain usage and multi-brand templates
  • Began using Cloudflare (AS13335) for hosting alongside existing Njalla (AS39287) and Virtuo (AS399486)
  • Registered domains targeting Klaviyo, HubSpot, Pure Storage, Morningstar, Instacart, Vodafone

Q2 2025 — UK Retail Attacks (April–May)

  • Marks & Spencer (April 2025): Attackers gained initial access as early as February 2025 by impersonating an M&S employee and calling the third-party service desk to obtain a password reset. Exfiltrated the NTDS.dit file from the Windows domain controller. Deployed DragonForce ransomware, halting warehouse operations for 46 days. Estimated £300M in lost profit; ~£1B stock valuation impact.
  • Co-op Group (May 2025): Attempted breach via similar social engineering; VPN access suspended as containment measure.
  • Harrods (May 2025): Intrusion contained; stores remained operational.
  • Additional targets: Dior, The North Face, Cartier, Victoria's Secret, Adidas at "unprecedented rate"

Q2–Q3 2025 — Aviation Sector Targeting (June–July)

  • Hawaiian Airlines (June 26, 2025): Cybersecurity event impacting IT systems
  • WestJet (June 13, 2025): Incident affecting internal systems and the WestJet app
  • Qantas (June 30, 2025): Unusual activity on third-party contact centre platform; customer data exposed (names, emails, phone numbers, birth dates, Frequent Flyer numbers)
  • FBI issued formal warning on June 30 about Scattered Spider targeting the airline industry

July 2025 — CISA/FBI Joint Advisory Update

  • Updated advisory AA23-320A published July 29, 2025
  • Added new TTPs: DragonForce ransomware, AnyDesk, Teleport.sh tunneling, RattyRAT, fake identity backstopping
  • Co-authored by Canadian Centre for Cyber Security, Australian Cyber Security Centre, and UK NCSC

Q3 2025–Q1 2026 — Continued Operations

  • Expanded targeting of SaaS platforms (Salesforce, ServiceNow, Workday)
  • Increased insider recruitment attempts
  • Continued high-volume AiTM phishing campaigns with automated spear-phishing tools

Tactics, Techniques, and Procedures (TTPs)

Initial Access

Social Engineering — Help Desk Exploitation (T1566.004, T1656)

The group's signature technique. Attackers impersonate employees or executives, calling IT help desks to request password resets or MFA device registration. They prepare by gathering employee information from LinkedIn, ZoomInfo, and breached databases. Callers are native English speakers with minimal accents, sometimes using regional dialects (e.g., Southern US). The M&S breach specifically used this vector against a third-party managed service desk.

AiTM Phishing with Evilginx (T1566, T1557)

Significantly escalated in 2025. Evilginx-based phishing pages capture both credentials and session tokens, bypassing MFA entirely. Notable behaviors:

  • Rickroll redirects: Evilginx configuration redirects security researchers to "Never Gonna Give You Up" YouTube video
  • Pre-populated victim information in phishing links
  • One-time magic links preventing security team analysis
  • Conditional content loading — login page only renders for specific usernames
  • Malvertising distribution via Google Ads to bypass email/network controls

SMS Phishing / Smishing (T1660)

High-volume campaigns using automated tools, some abusing Google Voice for identity harvesting at scale.

SIM Swapping (T1451)

Transferring victim phone numbers to attacker-controlled SIMs to intercept OTPs and MFA codes.

Trusted Relationship Exploitation (T1199)

Compromising MSPs and IT contractors (e.g., Tata Consultancy Services in the M&S attack) as a "one-to-many" access vector.

Execution & Persistence

Remote Monitoring & Management Tools (T1219)

Deploys multiple RMM tools for persistent access:

  • AnyDesk (newly added in 2025)
  • ScreenConnect / ConnectWise
  • Fleetdeck.io
  • Level.io
  • Pulseway
  • Splashtop
  • Tactical.RMM
  • TeamViewer

Tunneling for C2 (T1090, T1572)

  • Ngrok: Internet tunneling, often tunneling RDP (port 3389)
  • Teleport.sh: Custom hostname masking (e.g., CUSTOMERNAME.teleport[.]sh)
  • Tailscale: VPN mesh tunneling

Identity Federation Abuse (T1484.002)

Adding federated identity providers to victim SSO infrastructure for persistent, unrestricted tenant access.

MFA Token Registration (T1556.006)

Registering attacker-controlled MFA devices after obtaining password reset via help desk social engineering.

Credential Access

NTDS.dit Extraction (T1003.003)

Exfiltrating Active Directory database for offline credential harvesting — confirmed in the M&S attack.

LSASS Memory Dumping (T1003.001)

Using Mimikatz for credential extraction from memory.

MFA Fatigue / Push Bombing (T1621)

Sending repeated MFA notification prompts until victim accepts.

Evilginx captures session tokens that bypass MFA entirely.

Lateral Movement & Discovery

Cloud Instance Abuse (T1578.002, T1021.007)

Creating new EC2 instances or accessing pre-existing cloud VMs for lateral movement.

Active Directory Enumeration (T1482, T1018)

Using AdFind for domain trust discovery and network mapping. Targeting VMware vCenter, backup infrastructure, and VPN concentrators.

DRSGetNCChanges / DCSync

Directory replication requests to extract credentials from domain controllers.

Exfiltration

Multi-Channel Exfiltration (T1567, T1567.002)

  • MEGA[.]NZ cloud storage
  • Amazon S3 buckets
  • Dropbox, FiveTran SaaS connectors
  • SSH transfers to Vultr infrastructure
  • Snowflake Data Cloud queries via ETL tools

Impact

DragonForce Ransomware (T1486)

Current ransomware of choice. Built from leaked LockBit 3.0 and Conti source code. Targets Windows, Linux, and VMware ESXi systems. Uses RSA + AES / ChaCha8 encryption. Pre-encryption behavior includes shadow copy deletion, event log clearing, backup destruction, and service stopping.


MITRE ATT&CK Mapping

TacticTechnique IDTechnique Name
ReconnaissanceT1589Gather Victim Identity Information
ReconnaissanceT1593.001Search Open Websites/Domains: Social Media
ReconnaissanceT1594Search Victim-Owned Websites
ReconnaissanceT1597.002Search Closed Sources: Purchase Technical Data
ReconnaissanceT1598Phishing for Information
ReconnaissanceT1598.004Phishing for Information: Spearphishing Voice
Resource DevelopmentT1583.001Acquire Infrastructure: Domains
Resource DevelopmentT1585.001Establish Accounts: Social Media Accounts
Initial AccessT1078.002Valid Accounts: Domain Accounts
Initial AccessT1199Trusted Relationship
Initial AccessT1566Phishing
Initial AccessT1566.004Phishing: Spearphishing Voice
Initial AccessT1660SMS Phishing
ExecutionT1204User Execution
ExecutionT1648Serverless Execution
PersistenceT1078Valid Accounts
PersistenceT1136Create Account
PersistenceT1484.002Domain Policy Modification: Trust Modification
PersistenceT1556.006Modify Authentication Process: MFA
Defense EvasionT1090Proxy
Defense EvasionT1562.001Impair Defenses: Disable or Modify Tools
Defense EvasionT1656Impersonation
Credential AccessT1003.001OS Credential Dumping: LSASS Memory
Credential AccessT1003.003OS Credential Dumping: NTDS
Credential AccessT1451SIM Card Swap
Credential AccessT1539Steal Web Session Cookie
Credential AccessT1552.001Unsecured Credentials: Credentials in Files
Credential AccessT1606Forge Web Credentials
Credential AccessT1621Multi-Factor Authentication Request Generation
DiscoveryT1018Remote System Discovery
DiscoveryT1046Network Service Discovery
DiscoveryT1082System Information Discovery
DiscoveryT1083File and Directory Discovery
DiscoveryT1213.002Data from Information Repositories: SharePoint
DiscoveryT1213.003Data from Information Repositories: Code Repositories
DiscoveryT1482Domain Trust Discovery
Lateral MovementT1021.001Remote Services: RDP
Lateral MovementT1021.002Remote Services: SMB/Windows Admin Shares
Lateral MovementT1021.007Remote Services: Cloud Services
Lateral MovementT1578.002Modify Cloud Compute Infrastructure: Create Cloud Instance
CollectionT1074Data Staged
CollectionT1114Email Collection
CollectionT1530Data from Cloud Storage
Command and ControlT1071.001Application Layer Protocol: Web Protocols
Command and ControlT1219Remote Access Software
Command and ControlT1572Protocol Tunneling
ExfiltrationT1567Exfiltration Over Web Service
ExfiltrationT1567.002Exfiltration Over Web Service: Exfiltration to Cloud Storage
ImpactT1486Data Encrypted for Impact
ImpactT1490Inhibit System Recovery
ImpactT1657Financial Theft

Indicators of Compromise

Phishing Domain Patterns (Defanged)

DomainTargetDate
klv1[.]it[.]comKlaviyoFeb 2025
corp-hubspot[.]comHubSpotQ1 2025
morningstar-okta[.]comMorningstarQ1 2025
pure-okta[.]comPure StorageQ1 2025
signin-nydig[.]comNYDIGQ1 2025
sso-instacart[.]comInstacartQ1 2025
sts-vodafone[.]comVodafoneQ1 2025
paxos-my-salesforce[.]comPaxosFeb 2025
twitter-okta[.]comX/TwitterOct 2024
targetsname-cms[.]comGeneric pattern2025
targetsname-helpdesk[.]comGeneric pattern2025
oktalogin-targetcompany[.]comGeneric pattern2025

Domain naming conventions (keywords): okta, sso, help, he1p, helpdesk, internal, connect, duo, vpn, mfa, support, servicedesk, corp, dev, workspace, ops, hr, cdn

TLD preferences: .com, .co, .net, .org

Registrars: NiceNIC, Hosting Concepts B.V., NameSilo, GoDaddy

Domain lifespan: Typically active less than 7 days; some as short as 5 minutes to 2 hours

Hosting Infrastructure (ASNs)

ASNProviderNotes
AS13335CloudflareNew in January 2025
AS39287ABSTRACT (Njalla)Finland; active since Oct 2024
AS399486VirtuoCanada; active since Aug 2024
AS14061DigitalOceanHistorical
AS20473AS-CHOOPA (Vultr)Historical; used for exfiltration

DragonForce Ransomware Hashes

TypeHashNotes
SHA-2561ccf8baf11427fae273ffed587b41c857fa2d8f3d3c6c0ddaa1fe4835f665ebaDragonForce sample
SHA-256dc7e706587d4897789cc4a5f7cccbb539646b58aa9c86272728c8c1e6ec2a529DragonForce sample
SHA-256410db536a57c511b0ccac2639e0eb3320f303fc5c90242379ab43364c51ef321DragonForce sample
SHA-256f5df98b344242c5eaad1fce421c640fadd71f7f21379d2bf7309001dfeb25972DragonForce sample
SHA-25624e8ef41ead6fc45d9a7ec2c306fd04373eaa93bbae0bd1551a10234574d0e07DragonForce sample
SHA-256b10129c175c007148dd4f5aff4d7fb61eb3e4b0ed4897fea6b33e90555f2b845DragonForce sample
SHA-256d67a475f72ca65fd1ac5fd3be2f1cce2db78ba074f54dc4c4738d374d0eb19c7DragonForce sample
MD53a6e2c775c9c1060c54a9a94e80d923aDragonForce sample
MD574a97d25595ad73129fa946dc3156cecDragonForce sample
MD57ceeb2208a50b1ef61fdec935d66e992DragonForce sample
MD5e67e7b8e0fb6baff4f25bb05dd5a5e21DragonForce sample
MD5770c1dc157226638f8ad1ac9669f4883DragonForce sample

Spectre RAT Indicators

TypeValueNotes
Config file89CC88Dynamic C2 configuration
Config file733949System information storage
C2 URI patternwber command structureHTTP-based C2 protocol
Binary32-bit and 64-bit Intel PEXOR-encoded strings
Toolnircmdc.exeLOLBin used by Spectre RAT

Malware Arsenal

MalwareTypeStatus
DragonForceRansomware (RaaS)Active — primary 2025
Spectre RATRemote Access TrojanActive — updated 2025
RattyRATJava-based RATActive — added July 2025
Raccoon Stealer (S1148)InfostealerActive
VIDAR StealerInfostealerActive
AveMaria/WarZone (S0670)RATHistorical
POORTRYKernel driver (BYOVD)Active
STONESTOPLoader for POORTRYActive
Cobalt Strike BeaconC2 FrameworkActive via DragonForce
SystemBCSOCKS5 proxy backdoorActive via DragonForce
Mimikatz (S0002)Credential dumperActive

Legitimate Tools Abused

AnyDesk, ScreenConnect, Fleetdeck.io, Level.io, Pulseway, Splashtop, Tactical.RMM, TeamViewer, Ngrok, Teleport.sh, Tailscale, SimpleHelp RMM, Rclone, AdFind, SoftPerfect Network Scanner

Exfiltration Destinations

  • MEGA[.]NZ
  • Amazon S3 (US-based)
  • Dropbox
  • FiveTran (SaaS connector)
  • Snowflake Data Cloud
  • Vultr-hosted infrastructure
  • BitLaunch (cryptocurrency-paid servers)

Communication Channels

TOR, Tox, encrypted messaging applications, Telegram, Discord


Law Enforcement Actions

2024 Federal Charges (November 20, 2024)

Five members charged by US DOJ:

  • Tyler Robert Buchanan (22, UK) — arrested in Spain; 27M in Bitcoin traced
  • Ahmed Hossam Eldin Elbadawy (23, US)
  • Noah Michael Urban (20, US) — sentenced to 10 years for SIM-swapping ($13M)
  • Evans Onyeaka Osiebo (20, US)
  • Joel Martin Evans (25, US) Charges: conspiracy to commit wire fraud, conspiracy, aggravated identity theft — up to 25 years each.

2025 UK Indictments & Arrests

  • Thalha Jubair (19, UK) — charged with 7 counts including fraud and money laundering for allegedly participating in 120+ network intrusions targeting 47 US entities
  • Owen Flowers (18, UK) — indicted for attacks on US and British companies
  • 17-year-old involved in MGM/Caesars breaches — surrendered in Las Vegas

July 2025 UK Retail Attack Arrests

Four arrested in connection with M&S/Co-op/Harrods attacks:

  • 20-year-old British woman (Staffordshire)
  • 19-year-old Latvian male (West Midlands)
  • 19-year-old British man (London)
  • 17-year-old British male (West Midlands)

International Coordination

Operations involved FBI, National Crime Agency (UK), Spanish police, Las Vegas Metro Police, Canadian RCMP, and Australian Federal Police.


Detection Rules

The following detection rules target key Scattered Spider TTPs and DragonForce ransomware indicators. Rules cover help desk social engineering indicators in Okta logs, suspicious RMM tool deployment, tunneling for C2, Active Directory credential extraction, DragonForce pre-encryption behavior, identity federation abuse, and file-level malware indicators. Log sources covered: Okta IdP, Windows process creation (Sysmon/EDR), Azure AD audit logs, and file scanning (YARA).

Sigma Rules

1. Help Desk MFA Bypass Social Engineering Indicators (Okta)

Status: Validated

yaml
title: Scattered Spider - Help Desk MFA Bypass Social Engineering Indicators
id: 8a3f2e1d-4b5c-6d7e-9f0a-1b2c3d4e5f6a
status: experimental
description: Detects patterns consistent with Scattered Spider help desk social engineering - multiple failed MFA attempts followed by MFA factor reset and successful authentication within a short timeframe.
references:
    - https://www.cisa.gov/news-events/cybersecurity-advisories/aa23-320a
    - https://pushsecurity.com/blog/scattered-spider-ttp-evolution-in-2025
author: Actioner Research Agent
date: 2026-04-07
tags:
    - attack.initial_access
    - attack.t1566.004
    - attack.credential_access
    - attack.t1621
    - attack.t1556.006
logsource:
    product: okta
    service: okta
detection:
    selection_mfa_deny:
        eventType: 'user.authentication.auth_via_mfa'
        outcome.result: 'FAILURE'
    selection_mfa_reset:
        eventType:
            - 'user.mfa.factor.deactivate'
            - 'user.mfa.factor.activate'
            - 'system.mfa.factor.deactivate'
    selection_password_reset:
        eventType:
            - 'user.account.reset_password'
            - 'user.lifecycle.password_reset'
    condition: selection_mfa_deny or selection_mfa_reset or selection_password_reset
falsepositives:
    - Legitimate help desk password resets
    - User self-service MFA enrollment
level: medium

2. Suspicious RMM Tool Execution (Windows)

Status: Validated

yaml
title: Scattered Spider - Suspicious RMM Tool Execution
id: 7b4e3f2a-5c6d-8e9f-0a1b-2c3d4e5f6a7b
status: experimental
description: Detects execution of remote monitoring and management tools commonly abused by Scattered Spider for persistence and command-and-control.
references:
    - https://www.cisa.gov/news-events/cybersecurity-advisories/aa23-320a
    - https://reliaquest.com/blog/scattered-spider-cyber-attacks-using-phishing-social-engineering-2025/
author: Actioner Research Agent
date: 2026-04-07
tags:
    - attack.command_and_control
    - attack.t1219
    - attack.persistence
    - attack.t1078
logsource:
    category: process_creation
    product: windows
detection:
    selection_rmm:
        Image|endswith:
            - '\AnyDesk.exe'
            - '\ScreenConnect.ClientService.exe'
            - '\ScreenConnect.WindowsClient.exe'
            - '\saborern.exe'
            - '\saborern64.exe'
            - '\Pulseway.exe'
            - '\fleetdeck-agent.exe'
            - '\level-agent.exe'
            - '\TeamViewer.exe'
            - '\TeamViewer_Service.exe'
            - '\tacticalrmm.exe'
            - '\meshagent.exe'
    filter_legitimate:
        ParentImage|endswith:
            - '\services.exe'
            - '\svchost.exe'
        Image|endswith:
            - '\TeamViewer_Service.exe'
    condition: selection_rmm and not filter_legitimate
falsepositives:
    - Legitimate IT administration using authorized RMM tools
    - Software deployment systems
level: medium

3. Ngrok or Teleport Tunneling Activity (Windows)

Status: Validated

yaml
title: Scattered Spider - Ngrok or Teleport Tunneling Activity
id: 6c5d4e3f-2a1b-9c8d-7e6f-5a4b3c2d1e0f
status: experimental
description: Detects execution of tunneling tools (Ngrok, Teleport) commonly used by Scattered Spider for command-and-control and to bypass network security controls.
references:
    - https://www.cisa.gov/news-events/cybersecurity-advisories/aa23-320a
    - https://www.darktrace.com/blog/untangling-the-web-darktraces-investigation-of-scattered-spiders-evolving-tactics
author: Actioner Research Agent
date: 2026-04-07
tags:
    - attack.command_and_control
    - attack.t1090
    - attack.t1572
logsource:
    category: process_creation
    product: windows
detection:
    selection_ngrok:
        Image|endswith: '\ngrok.exe'
    selection_teleport:
        Image|endswith: '\tsh.exe'
    selection_cmdline_ngrok:
        CommandLine|contains:
            - 'ngrok'
            - 'tcp 3389'
            - 'tcp 445'
    selection_cmdline_teleport:
        CommandLine|contains:
            - 'teleport.sh'
            - 'tsh login'
            - 'tsh ssh'
    condition: selection_ngrok or selection_teleport or selection_cmdline_ngrok or selection_cmdline_teleport
falsepositives:
    - Legitimate developer use of ngrok for testing
    - Authorized Teleport deployments
level: high

4. NTDS.dit Credential Extraction (Windows)

Status: Validated

yaml
title: Scattered Spider - NTDS.dit Credential Extraction
id: 5d6e7f8a-9b0c-1d2e-3f4a-5b6c7d8e9f0a
status: experimental
description: Detects attempts to access or copy the NTDS.dit Active Directory database file, a technique used by Scattered Spider and DragonForce affiliates to harvest domain credentials before ransomware deployment.
references:
    - https://www.cisa.gov/news-events/cybersecurity-advisories/aa23-320a
    - https://specopssoft.com/blog/marks-spencer-ransomware-active-directory/
author: Actioner Research Agent
date: 2026-04-07
tags:
    - attack.credential_access
    - attack.t1003.003
    - attack.t1003.001
logsource:
    category: process_creation
    product: windows
detection:
    selection_ntdsutil:
        Image|endswith: '\ntdsutil.exe'
        CommandLine|contains:
            - 'ifm'
            - 'create full'
            - 'ac i ntds'
    selection_vssadmin:
        Image|endswith: '\vssadmin.exe'
        CommandLine|contains: 'create shadow'
    selection_copy_ntds:
        CommandLine|contains:
            - 'ntds.dit'
            - 'NTDS.dit'
    selection_secretsdump:
        CommandLine|contains:
            - 'secretsdump'
            - 'impacket'
    condition: selection_ntdsutil or selection_vssadmin or selection_copy_ntds or selection_secretsdump
falsepositives:
    - Legitimate Active Directory backup operations
    - Authorized security assessments
level: critical

5. DragonForce Ransomware Pre-Encryption Indicators (Windows)

Status: Validated

yaml
title: DragonForce Ransomware - Pre-Encryption Indicators
id: 4e5f6a7b-8c9d-0e1f-2a3b-4c5d6e7f8a9b
status: experimental
description: Detects pre-encryption behavior associated with DragonForce ransomware, including shadow copy deletion, event log clearing, and service stopping - techniques observed in Scattered Spider affiliate operations.
references:
    - https://www.picussecurity.com/resource/blog/dragonforce-ransomware-attacks-retail-giants
    - https://www.cisa.gov/news-events/cybersecurity-advisories/aa23-320a
author: Actioner Research Agent
date: 2026-04-07
tags:
    - attack.impact
    - attack.t1490
    - attack.t1486
    - attack.defense_evasion
    - attack.t1070.001
logsource:
    category: process_creation
    product: windows
detection:
    selection_shadow_delete:
        Image|endswith:
            - '\vssadmin.exe'
            - '\wmic.exe'
        CommandLine|contains:
            - 'delete shadows'
            - 'shadowcopy delete'
    selection_bcdedit:
        Image|endswith: '\bcdedit.exe'
        CommandLine|contains:
            - 'recoveryenabled no'
            - 'bootstatuspolicy ignoreallfailures'
    selection_wevtutil:
        Image|endswith: '\wevtutil.exe'
        CommandLine|contains:
            - 'cl System'
            - 'cl Security'
            - 'cl Application'
    selection_service_stop:
        Image|endswith: '\net.exe'
        CommandLine|contains:
            - 'stop "SQLServerAgent"'
            - 'stop "MSSQLSERVER"'
            - 'stop "vss"'
            - 'stop "backup"'
    condition: 1 of selection_*
falsepositives:
    - System administrators performing maintenance
    - Legitimate backup software managing shadow copies
level: critical

6. Suspicious Identity Federation Configuration Change (Azure AD)

Status: Validated

yaml
title: Scattered Spider - Suspicious Identity Federation Configuration Change
id: 3f4a5b6c-7d8e-9f0a-1b2c-3d4e5f6a7b8c
status: experimental
description: Detects addition of new federated identity providers or domain federation changes in Azure AD/Entra ID, a persistence technique used by Scattered Spider to maintain access through SSO abuse.
references:
    - https://www.cisa.gov/news-events/cybersecurity-advisories/aa23-320a
    - https://pushsecurity.com/blog/scattered-spider-ttp-evolution-in-2025
author: Actioner Research Agent
date: 2026-04-07
tags:
    - attack.persistence
    - attack.t1484.002
    - attack.privilege_escalation
logsource:
    product: azure
    service: auditlogs
detection:
    selection_federation:
        operationName:
            - 'Set domain authentication'
            - 'Set federation settings on domain'
            - 'Add domain to company'
            - 'Set company information'
    selection_idp:
        operationName:
            - 'Add identity provider'
            - 'Update identity provider'
    condition: selection_federation or selection_idp
falsepositives:
    - Legitimate federation configuration by IT administrators
    - Planned SSO integration projects
level: high

YARA Rules

7. DragonForce Ransomware Indicators

Status: Validated

yara
import "pe"

rule DragonForce_Ransomware_Indicators
{
    meta:
        description = "Detects DragonForce ransomware based on known string patterns and behavioral indicators from LockBit3/Conti lineage"
        author = "Actioner Research Agent"
        date = "2026-04-07"
        reference = "https://www.picussecurity.com/resource/blog/dragonforce-ransomware-attacks-retail-giants"
        severity = "critical"
        hash1 = "1ccf8baf11427fae273ffed587b41c857fa2d8f3d3c6c0ddaa1fe4835f665eba"
        hash2 = "dc7e706587d4897789cc4a5f7cccbb539646b58aa9c86272728c8c1e6ec2a529"

    strings:
        $ransom_note1 = "DragonForce" ascii wide nocase
        $ransom_note2 = "YOUR FILES HAVE BEEN ENCRYPTED" ascii wide nocase
        $ransom_note3 = ".dragonforce" ascii wide nocase
        $lockbit_heritage1 = "LockBit" ascii wide
        $lockbit_heritage2 = "lockbit3" ascii wide nocase

        $cmd_shadow1 = "vssadmin delete shadows /all /quiet" ascii wide nocase
        $cmd_shadow2 = "wmic shadowcopy delete" ascii wide nocase
        $cmd_bcdedit = "bcdedit /set {default} recoveryenabled no" ascii wide nocase
        $cmd_wevtutil = "wevtutil cl" ascii wide

        $mutex1 = "Global\\DragonForce" ascii wide
        $ext1 = ".dragonforce_encrypted" ascii wide

    condition:
        uint16(0) == 0x5A4D and
        filesize < 5MB and
        (
            (2 of ($ransom_note*)) or
            ($mutex1) or
            ($ext1 and 1 of ($cmd_*)) or
            (2 of ($cmd_*) and 1 of ($ransom_note*)) or
            (1 of ($lockbit_heritage*) and 1 of ($ransom_note*) and 1 of ($cmd_*))
        )
}

8. Spectre RAT (Scattered Spider)

Status: Validated

yara
rule Spectre_RAT_Scattered_Spider
{
    meta:
        description = "Detects Spectre RAT malware associated with Scattered Spider operations - XOR-encoded strings and C2 configuration patterns"
        author = "Actioner Research Agent"
        date = "2026-04-07"
        reference = "https://www.silentpush.com/blog/scattered-spider-2025/"
        severity = "high"

    strings:
        $config1 = "89CC88" ascii wide
        $config2 = "733949" ascii wide
        $wber = "wber" ascii
        $cmd_exec = "cmd.exe /c" ascii wide
        $nircmd = "nircmdc.exe" ascii wide nocase
        $mutex_check = "CreateMutex" ascii
        $http_ua = "Mozilla/5.0" ascii
        $xor_loop = { 8B ?? 33 ?? 89 ?? 4? FF ?? 75 }

    condition:
        uint16(0) == 0x5A4D and
        filesize < 3MB and
        (
            ($config1 and $config2) or
            ($wber and $mutex_check and $http_ua) or
            ($config1 and $xor_loop) or
            ($nircmd and 2 of ($config1, $config2, $wber, $cmd_exec))
        )
}

Detection Rules Summary

#RuleTypeTargetsLevelStatus
1Help Desk MFA BypassSigma (Okta)T1566.004, T1621, T1556.006MediumValidated
2Suspicious RMM Tool ExecutionSigma (Windows)T1219, T1078MediumValidated
3Ngrok/Teleport TunnelingSigma (Windows)T1090, T1572HighValidated
4NTDS.dit Credential ExtractionSigma (Windows)T1003.003, T1003.001CriticalValidated
5DragonForce Pre-EncryptionSigma (Windows)T1490, T1486, T1070.001CriticalValidated
6Federation Configuration ChangeSigma (Azure AD)T1484.002HighValidated
7DragonForce RansomwareYARAFile-level detectionCriticalValidated
8Spectre RATYARAFile-level detectionHighValidated

Defensive Recommendations

Immediate Priority

  1. Deploy phishing-resistant MFA — FIDO2/WebAuthn or PKI-based; eliminates push bombing, SIM swap, and AiTM attacks
  2. Harden help desk identity verification — Implement callback verification, video confirmation, or manager-approved identity checks before password resets or MFA changes
  3. Audit and allowlist RMM tools — Block all unauthorized remote access software; alert on any RMM execution not in the approved list
  4. Monitor for NTDS.dit access — Deploy Sigma rule #4; any access to NTDS.dit outside scheduled AD backups is critical
  5. Protect domain controllers — Enable LSASS protection, credential guard, and DCSync detection

Network & Infrastructure

  1. Block tunneling services — Egress filter for ngrok[.]io, teleport[.]sh, tailscale domains; deploy Sigma rule #3
  2. Segment networks — Isolate backup infrastructure, domain controllers, and VMware management from general user networks
  3. Monitor cloud infrastructure — Alert on new EC2 instances, cloud storage access from unusual sources, and federation configuration changes
  4. Implement immutable backups — Offline, air-gapped, regularly tested backup copies

Identity & Access

  1. Restrict third-party access — Harden MSP/contractor access with dedicated jumpboxes, mandatory MFA, and limited session duration
  2. Monitor Okta/Entra ID logs — Deploy Sigma rules #1 and #6; alert on MFA factor changes, federation modifications, and bulk password resets
  3. Implement conditional access — Risk-based authentication with impossible travel detection, device compliance, and anomalous login alerting

Employee Training

  1. Vishing awareness — Train help desk staff on social engineering pretexts; simulate attacks
  2. OTP/MFA code sharing — Strict policy against sharing authentication codes under any circumstances
  3. Report suspicious calls — Clear escalation path for help desk personnel when receiving unusual reset requests

Sources

MIT License