← Breach analysis by the Dual Vulnerabilities framework

Russian state institutions espionage phishing (Kasablanka Group, 2022)

2022 · Russia · O - Public administration and defence · State-sponsored espionage via targeted phishing with RAT payloads

Analysed 2026-11-06 by Hermes

Open this case in the tool →

What happened

State-sponsored espionage via targeted phishing with RAT payloads

Dual vulnerability type

No DV evidenced

Alternative: none defensible

Why this classification

(1) The strategic expectation at both victim institutions (Rossotrudnichestvo and the Astrakhan Region ministry responsible for public administration and IT/communications) would be to keep malicious documents out of mailboxes and to notice and evict a remote-access trojan; nothing in the sources documents what either institution had decided, resourced or communicated on these controls. (2) What is documented is entirely attacker-side: Qi-Anxin's January 2023 report analysed captured samples of a September-December 2022 phishing campaign against Russian state institutions, in which socially engineered phishing emails carried virtual-disk (.vhdx) attachments nesting LNK downloaders that ultimately ran Warzone RAT and then Loda RAT, with some samples showing a zero detection rate on VirusTotal; the campaign is assessed as information-gathering espionage. Cisco Talos's March 2023 report on the overlapping YoroTrooper clusters corroborates the same September-December 2022 Russia-targeted VHDX/LodaRAT activity and notes Kasablanka is the developer but not the sole operator of LodaRAT. (3) No victim-side lifecycle stage yields a mechanism: neither institution made any statement or disclosure (the incident was disclosed by an IT-security company months after the campaign, consistent with the EuRepoC record), no detection or dwell time, containment, data exfiltration impact, recovery or response conduct is reported in English or Russian sources, and EuRepoC codes the incident as hijacking without misuse - access obtained but not used, no system interference, zero economic impact. The entry vector is known (spear phishing) but is the attacker's move, not a victim-side condition: no source shows a training, verification, filtering or procedure failure at either victim, so reading a DV-Type off the phishing entry or the zero sample-detection rate would be the rejected attacker-technique reasoning. Reporting here confirms targeting only, which is the decision rule's case for Unknown; the case is also atypical in the framework's sense, a Morocco-based espionage operation against Russian state bodies documented solely by security vendors. Risk markers and the stage distribution are left blank accordingly.

Sources

Notes

Attribution rests on Qi-Anxin with medium (moderate) confidence and the report itself does not rule out a false-flag attempt via reverse-engineered Loda RAT source; Cisco Talos likewise finds Kasablanka is the developer but not the sole operator of LodaRAT. Victim-side silence: neither Rossotrudnichestvo nor the Astrakhan Region ministry acknowledged the campaign; searches in English and Russian (agency name + кибератака/утечка) surfaced no victim statement, regulator notice or impact report. Data quality: the case file's Victim column names the Ministry of Public Administration, IT and Communications of Astrakhan Region, while the row's own description and the Qi-Anxin report call the target the Ministry of Foreign Communications of the Astrakhan Region; the victim is kept as written in the case file. The case file's third victim position is Not available and is not recorded. The twitter.com source_url is behind a login wall and could not be opened, so it is not cited. No related_rows in the case file.