Join our Newsletter — 33% off our NHI Course
Home› Guides› Blind Eagle Threat Actor Profile Guide: Colombian government…
Guide Threat Intelligence & Threat Actors

Blind Eagle Threat Actor Profile Guide: Colombian government and bank targeting explained

← All guides
By NHI Mgmt Group, edited by Lalit Choda Updated 27 October 2026 10 min read
On this page

Blind Eagle (APT-C-36) is the most persistent targeted threat to Colombian government bodies and banks, and it does it with commodity remote access trojans, compromised government mailboxes and lures that look like tax, court and prosecutor notices. It switches between espionage and financial credential theft, sometimes within the same period, which makes it hard to place in the usual state or criminal box. This profile is for SOC analysts, email security teams, fraud teams and intelligence staff in Colombia, Ecuador and the wider Latin American region, and for any organisation that handles Spanish-language mail from those governments. It covers names and attribution as each source states them, targeting, ATT&CK techniques, how the group steals and reuses identities, notable campaigns and a detection and hardening plan. General background is in the Threat Actor Groups Guide.

Key takeaways

  • It lives off trusted senders. ATT&CK and Recorded Future report regular use of compromised email accounts, including Colombian government and judicial addresses, and a compromised regional government Google Drive used to host payloads.
  • Lures impersonate authority. Kaspersky lists DIAN (the tax and customs authority), the Ministry of Foreign Affairs, the Attorney General's Office and judicial institutions; phishing pages copy Bancolombia, BBVA and Davivienda.
  • The tooling is cheap but varied. Modified njRAT, AsyncRAT, Remcos, DcRAT, LimeRAT and Quasar variants, packed with crypters such as HeartCrypt and PureCrypter.
  • It is fast to adopt new tricks. Check Point saw a .url file technique related to CVE-2024-43451 six days after Microsoft patched it in November 2024; one December 2024 campaign reached more than 1,600 victims.
  • Geofencing hides it from outsiders. Delivery links redirect anyone outside Colombia or Ecuador to the real impersonated site, so scanning from abroad often shows nothing.

Who they are (names and aliases)

MITRE ATT&CK tracks the group as APT-C-36 (G0099), a suspected South American threat group that has engaged in espionage and financially motivated operations since at least 2018, targeting government institutions and the financial, energy and manufacturing sectors across Colombia and other Latin American countries. Check Point says the group has long been suspected to originate from South America and noted repository commit timestamps in UTC-5. Recorded Future states that the group's origin remains uncertain, that some research points to Colombia, and that its use of Brazilian criminal tools suggests a possible link to Brazilian actors. No government attribution has been published in the sources reviewed.

NameTracked byNotes
APT-C-36Qihoo 360 / QiAnXin (2019), ATT&CKFirst public report described attacks on Colombian government and companies
Blind Eagle, AguilaCiegaKaspersky, Check Point; listed in ATT&CKMost common public name in English and Spanish
TAG-144Recorded Future Insikt GroupFive activity clusters from 2024 to 2025; linked to the Red Akodon cluster
APT-Q-98Listed in ATT&CKVendor alias

Targeting and motivation

  • Colombia first: Kaspersky found Colombia accounted for 87% of victims in its May and June 2024 espionage campaigns. Recorded Future describes victims at local, municipal and federal government level.
  • Justice system: Check Point's December 2024 and January 2025 lures pointed mainly at the Colombian justice system.
  • Region: Ecuador, Chile and Panama also appear in Kaspersky's reporting.
  • Sectors: government, banking and finance, energy and oil and gas, with education, health, transport, retail and defence also hit (Kaspersky, Recorded Future).
  • Motivation: both. Kaspersky saw modified njRAT used for espionage and a modified Quasar used as a banking trojan; Recorded Future says it is unclear whether the activity is financially motivated, state-sponsored espionage or both.

Tactics and techniques (MITRE ATT&CK)

All techniques below are mapped to G0099 in ATT&CK v20. Software mapped to the group includes njRAT (S0385), AsyncRAT (S1087), Remcos (S0332), QuasarRAT (S0262, including the BlotchyQuasar variant), Imminent Monitor (S0434), DCRAT (S9017), Caminho (S9016), PureCrypter (S9019) and HeartCrypt (S9018).

TacticTechniqueHow it shows up
ReconnaissanceT1593 Search Open Websites/DomainsResearch on Colombian banks to build phishing pages
Resource developmentT1586.002 Email Accounts; T1586.003 Cloud AccountsCompromised mailboxes and a government-linked Google Drive
Resource developmentT1583.006 Web Services; T1608.001 Upload MalwarePayloads on GitHub, Bitbucket, Google Drive, Dropbox, Discord, Pastebin and image hosts
Resource developmentT1683.001 Written Content; T1684.001 ImpersonationNotices that copy DIAN, prosecutors, courts and banks
Initial accessT1566.001 Spearphishing Attachment; T1566.002 Spearphishing LinkPDF and DOCX lures, password-protected RAR files, shortened links
Lateral movementT1534 Internal SpearphishingA compromised account mailing an address likely monitored by the victim's IT team
ExecutionT1059.005 Visual Basic; T1059.007 JavaScript; T1059.001 PowerShell; T1047 WMIScript chains, PowerShell started through WMI
PersistenceT1053.005 Scheduled Task; T1036.004 Masquerade Task or ServiceTasks disguised as Google tasks
StealthT1480 Execution GuardrailsGeolocation filtering that sends non-targets to legitimate sites
StealthT1027.003 Steganography; T1027.013 Encrypted/Encoded File; T1564.003 Hidden WindowCode hidden in images; hidden PowerShell windows
Stealth, privilege escalationT1055.012 Process Hollowing; T1574.001 DLL side-loadingInjection into legitimate processes; HijackLoader side-loaded
Command and controlT1568 Dynamic Resolution; T1571 Non-Standard PortDuckDNS, No-IP and con-ip domains; C2 on port 4050

How they abuse identities and credentials

Identity is central to Blind Eagle on both sides of the attack: it borrows the identity of government senders to get in, and it steals banking and system credentials once it is there.

Identity abusedWhat the actor doesControl that limits it
Government mailboxesPhishing sent from compromised mayor's office and judicial-branch accounts (Recorded Future)Phishing-resistant MFA for public sector mail, mailbox audit logging, inbound scanning that does not trust a sender because the domain is .gov.co
Cloud storage accountsPayloads hosted in a compromised regional government Google Drive (ATT&CK, Recorded Future)Treat shared-drive download links in mail as untrusted; report abuse quickly to the provider
Online banking credentialsModified Quasar starts keylogging when a browser title matches one of ten Colombian financial entities (Kaspersky); fake bank login portalsBank-side device binding and transaction signing; customer anti-phishing messaging; monitoring of lookalike domains
Internal IT trustMail from a compromised account to an address watched by the IT team (ATT&CK)Treat internal mail with attachments as untrusted; flag mail to shared IT mailboxes with archives
Remote sessionsModified njRAT can create RDP sessions (Kaspersky)Restrict RDP to managed jump hosts, enforce Network Level Authentication and MFA
NTLM authentication.url files that make WebDAV requests when a file is touched; Check Point says SMB signing and NTLM relay protections matter for the original CVEBlock outbound SMB and WebDAV to the internet, enforce SMB signing, patch CVE-2024-43451

For the NTLM side, see the NTLM Relay Guide and the SMB Signing Guide. For mailbox takeover response, see the Account Takeover Response Playbook.

Notable incidents

DateActivitySource
2018 onwardAttacks on Colombian government institutions and companiesQihoo 360 / QiAnXin, February 2019, via ATT&CK
May to June 2024Espionage campaigns with DIAN and judicial lures; HijackLoader and AsyncRAT; Portuguese strings in one campaignKaspersky, 19 August 2024
May 2024 to July 2025Five TAG-144 clusters using AsyncRAT, Remcos, DcRAT and LimeRAT against Colombian government entitiesRecorded Future, August 2025
November 2024.url technique related to CVE-2024-43451 seen six days after the patchCheck Point, 10 March 2025
December 2024More than 1,600 victims in one campaign and about 9,000 infections across December campaignsCheck Point
February 2025Actor briefly posted stolen phishing data to its own GitHub repositoryCheck Point

Detection and defence

Detection signals

SignalWhere to lookNotes
Spanish-language legal, tax or court notices with links to Google Drive, Dropbox, GitHub or BitbucketMail gateway, URL detonationDetonate from a Colombian egress point if possible, because of geofencing
Password-protected archives with the password in the email bodyMail gatewayUsed to get past attachment scanning
.url files in downloads or mail, and WebDAV requests from explorer.exeEDR file and network events, proxy logsThe CVE-2024-43451 related technique
wscript.exe or PowerShell launched by WMI with hidden windowsSysmon Event ID 1, WMI activity logsScript chain
Scheduled tasks named like Google updaters that point to user-writable pathsEvent ID 4698, autoruns baselinesMasquerading persistence
Endpoints resolving DuckDNS, No-IP or con-ip domainsDNS logs, protective DNSBlock where not needed
Outbound connections on port 4050 or other unusual portsFirewall and NetFlowRAT C2
Mailbox sending to unusual internal or external recipients after a sign-in from a VPN providerIdentity provider and mail audit logsCompromised sender accounts

Hardening plan

  1. Week 1: load current Kaspersky, Check Point and Recorded Future indicators into DNS and mail filters, and block dynamic DNS domains at the resolver where the business does not need them. See the Protective DNS Guide.
  2. Weeks 2 to 4: quarantine password-protected archives and .url, .vbs, .js and .lnk attachments at the gateway; patch CVE-2024-43451 (in the CISA KEV catalogue) and block outbound SMB and WebDAV.
  3. Weeks 4 to 8: apply application control or attack surface reduction rules to stop script hosts launched from mail and downloads, in audit mode first, then enforce by department with a documented rollback.
  4. Quarter 1: enforce phishing-resistant MFA on government and partner-facing mailboxes, enable mailbox audit logs and alert on new inbox rules and unusual sending. Banks should add transaction signing and lookalike domain takedown.
  5. Response: isolate, reimage, rotate the user's banking, mail and VPN credentials, and check whether the mailbox was then used to phish others. Tell partners if your domain sent lures.

Metrics

  • Password-protected archives delivered to users per month: zero outside approved senders.
  • Endpoints resolving dynamic DNS domains: trending to zero.
  • Government-facing mailboxes on phishing-resistant MFA: 100%.
  • Time from compromised sender report to mailbox lockdown: under one hour.
  • Script host launches from mail or download folders blocked: 100% after enforcement.

Framework mapping

FrameworkReferenceRelevance
NIST SP 800-53 Rev. 5SI-8 Spam Protection, SI-3 Malicious Code ProtectionMail filtering of lures and archives
NIST SP 800-53 Rev. 5IA-2 Identification and Authentication, AC-2 Account ManagementMailbox MFA and compromised account handling
NIST SP 800-53 Rev. 5SC-7 Boundary Protection, SI-4 System MonitoringBlocking outbound SMB, WebDAV and dynamic DNS
NIST CSF 2.0PR.AT-01, DE.CM-09, DE.AE-02Awareness, endpoint monitoring, alert analysis
MITRE D3FENDD3-DNSDL DNS Denylisting, D3-EAL Executable AllowlistingBlocking C2 and unwanted script execution

Practitioner checklist

  • Do not trust Colombian government sender domains by default; scan their mail as strictly as any other.
  • Quarantine password-protected archives and script or shortcut attachments.
  • Patch CVE-2024-43451 and block outbound SMB and WebDAV.
  • Block dynamic DNS domains and alert on RAT-style ports such as 4050.
  • Hunt for scheduled tasks masquerading as Google updaters.
  • Put phishing-resistant MFA on public sector and finance mailboxes.
  • After an infection, rotate banking and mail credentials and check for onward phishing.
  • Detonate suspicious links from a regional egress point so geofenced payloads are seen.

Standards and references

Related NHI Mgmt Group resources: Threat Actor Groups Guide · Latin American Banking Trojans Guide · AsyncRAT, Remcos and njRAT Commodity RATs Guide · Steganography in Attacks Guide · Account Takeover Prevention Guide

Free weekly newsletter

Subscribe to the NHI & AI Identity Journal

The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.

Bonus 33% off our NHI Course when you subscribe.

NHIMG Editorial Note
Researched and written by the NHI Mgmt Group editorial team using AI tools, under the editorial direction of Lalit Choda. Facts are drawn from the sources listed on this page and checked by our automated quality process. Last updated 27 October 2026. Spotted an error? Use our contact page.
NHI Mgmt Group, the independent authority on Non-Human Identity, IAM and Agentic AI security. nhimg.org