← Breach analysis by the Dual Vulnerabilities framework

Ukrainian & Polish government data theft (WinterVivern, 2023)

2023 · Poland, Ukraine · O - Public administration and defence · Data theft via phishing and fake government websites

Analysed 2026-11-06 by Hermes

Open this case in the tool →

What happened

Data theft via phishing and fake government websites

Dual vulnerability type

Strategy to Execution

Evidence coverage: 1/6 stages, weight 8/60

Decisive stage: entry (clearest evidence) · Alternative: Expertise to Execution

Why this classification

(1) The victims are government institutions of Ukraine and Poland, under continuous state-linked cyber attack, where the expectation of caution and source-verification against exactly this lure was established knowledge at the victim states' strategic level: CERT-UA had tracked UAC-0114's fake-government-website, fake-malware-scan technique since June 2022, SCPC confirmed in its own report that the same technique was used in the 2022 attack on the Ministry of Defence mail portal, and CBZC published the rule that police units publish no diagnostic software on their websites and that software must come only from official producer domains. (2) In practice that guidance did not reach ministry operations: on 30 January 2023 an email impersonating the MFA's own IT security (sent from outside the organisation) reached the corporate mailbox, staff downloaded and ran Protector.bat from a fake MFA page, PowerShell payloads executed unhindered, took screenshots, enumerated desktop documents and exfiltrated them over HTTP via the APERETIF implant, and Polish institutions received the identical lure through a fake CBZC page. (3) That is a Strategy to Execution gap: the knowledge and expectation existed doctrinally (repeated national-level warnings about a known lure, published verification norms), no source shows leadership deferring, under-resourcing or accepting the control, and what is missing is the translation of the guidance into ministry-level routines and operating practice, so the chain broke at execution. The classification rests on the entry stage, the only stage with a documented victim-side gap: detection and dwell time are undisclosed, egress controls and containment/blast radius are undocumented, recovery is not in play in an exfiltration-only case, and the state CERT response was aligned; the entry facts are documented by primary sources, and the origin and break readings are the coder's inference from them. (4) RM24 (Strategy to Execution lens): verification and caution requirements about downloading scan software were not translated into effective procedures at the institutions - staff obeyed an external email posing as their own IT security and executed software downloaded from a spoofed government website, even though the lure theme had been publicly warned about (causal role: Enabling). RM21 (Strategy to Execution lens): the lesson of the earlier comparable attack did not become a durable change - CERT-UA's own alert records that the identical fake-website, fake-scanner technique succeeded against a Ukrainian government institution in June 2022 and succeeded again in January 2023 (causal role: Perpetuating).

Evidence by stage

Risk markers

Sources

Notes

Link check: scpc.gov.ua blocks automated access (HTTP 403); the page was opened and read during research. The dataset source_url and attribution_source_url (a scpc.gov.ua api/docs PDF) return an edge-firewall Access Denied page when opened; the SCPC's own English report page for the same February 2023 report (articles/231) and the CERT-UA alert pages (cert.gov.ua 3761023 and 3761104) were opened and cited instead. The thehackernews.com source_url is the group's separate March 2023 Zimbra-exploitation reporting and was not used to classify this January 2023 phishing campaign; the two twitter.com source_urls were not retrieved. Victim is a dataset category (no receiver_name); the named institutions come from the primary sources. The earlier comparable incident (June 2022) hit the Ministry of Defence while the 2023 targets include the MFA - different ministries within the same victim category (government institutions), so RM21 is coded at that category level, matching the row's own victim framing. The entry stage's origin and break answers are the coder's reading: the primary sources document the infection chain, the prior campaign and the published verification norm, but state nothing about the hit institutions' internal controls or leadership decisions. Country records both victim states as the dataset does (receiver_country Poland;Ukraine). Row description and victim category match; no data-quality mismatch found. Unattended run: nothing requiring consent was attempted.