← Breach analysis by the Dual Vulnerabilities framework

Russian government organizations and IT companies EastWind espionage campaign (APT31, 2024)

2024 · Russia · O - Public administration and defence · Cyber espionage via spear-phishing and cloud-based backdoors

Analysed 2026-11-06 by Hermes

Open this case in the tool →

What happened

Cyber espionage via spear-phishing and cloud-based backdoors

Dual vulnerability type

No DV evidenced

Alternative: none defensible

Why this classification

(1) Strategic expectation at the victim: the targeted entities are Russian government organisations (receiver_subcategory: government/ministries) and IT companies - categories expected to hold security policies and monitoring able to resist targeted espionage - but every victim is unnamed in all reporting and no source documents any policy, programme, control or decision at any affected entity, so no doctrinal or material expectation can be evidenced. (2) What happened in practice: in late July 2024 Kaspersky detected spear-phishing emails carrying RAR archives with malicious LNK shortcuts that infected dozens of computers at Russian government organisations and IT companies; clicking the shortcut DLL-side-loaded a Dropbox-commanded backdoor, which pulled further payloads - the APT31 RAT GrewApacha, an updated CloudSorcerer backdoor and the previously unknown PlugY implant - with C2 addresses hidden in GitHub, Dropbox, Quora, LiveJournal and Yandex.Disk profiles. The campaign was discovered by the IT-security vendor, not by the victims; none of the affected organisations made any statement, and per Newsweek the Russian and Chinese foreign ministries did not respond to a written request for comment. (3) Why this is Unknown rather than a DV-Type: the framework requires a concrete, verifiable victim-side failure point; here the reporting confirms impact and attacker TTPs only. No victim organisation is named, and nothing is documented about any victim's own detection timeline, containment, egress scoping, recovery or response, so the two DV questions (origin and break) cannot be answered from evidence at any stage. Reading the vendor-side discovery as a monitoring failure, or the successful phishing as an awareness gap, would be the coder's own inference - the RM24 mechanism (inadequate training, verification or procedure at a named organisation) is nowhere evidenced - and the method excludes such inference as a basis; the entry vector is the attacker's move, not a victim-side mechanism. Follow-up reporting (Positive Technologies, November 2025, on a related APT31 cluster against Russian IT contractors, overlapping EastWind) documents long undetected dwell times and exfiltration via Yandex Cloud but likewise names no organisation and no organisational condition; it is actor-pattern context only and does not lift this row to a DV-Type.

Sources

Notes

Data quality: eurepoc_extra.intelligence_impact ('No data breach/exfiltration...') is EuRepoC's coding of observed impact and reads oddly for a coded 'Hijacking with Misuse' espionage case - Kaspersky's report documents an UPLOAD command in the Dropbox backdoor and keylogging/screen capture, but no public source ties a specific data exfiltration event to this campaign, so the coding was kept as-is. Victim naming: the dataset carries no receiver_name; victims are 'Russian government organizations and IT companies' (dozens of computers), unnamed in every source; no Russian CERT/GKRRN, ministry or victim statement was found (Newsweek, 13-08-2024: Russian and Chinese foreign ministries did not respond to request for comment). Disclosure: by an IT-security company (Kaspersky, Securelist report 08-08-2024), not by any victim. Follow-up context only: Positive Technologies (reported 24-11-2025) describes APT31 intrusions in Russian IT companies serving government agencies during 2024-2025 with overlaps to EastWind and dwell times of up to years; it names no victim organisation and documents no victim-side control or decision, so it was not used as lifecycle evidence. Sector: mixed victim set (receiver_subcategory: Government/ministries x3, Not available x3 for the IT companies); coded 'O Public administration and defence' following the government-dominated receiver categories. Set to Unknown because the sources confirm impact and attacker TTPs only and no lifecycle stage yields a documented victim-side mechanism at a nameable organisation.