← Breach analysis by the Dual Vulnerabilities framework

Stanton Water Department HMI manipulation (People's Cyber Army, 2024)

2024 · United States · E - Water supply · OT manipulation via internet-exposed HMI

Analysed 2026-11-06 by Hermes

Open this case in the tool →

What happened

OT manipulation via internet-exposed HMI

Dual vulnerability type

Expertise to Execution

Evidence coverage: 3/6 stages, weight 18/60

Decisive stage: detection (clearest evidence) · Alternative: Strategy to Leadership

Why this classification

(1) Stanton is a municipally run drinking-water utility serving 2,700 people - a designated critical-infrastructure operator whose control layer (the HMI that sets filtration, turbidity and flow) was expected to be unreachable from the internet and whose remote manipulation was expected to be noticed: EPA and CISA published exactly these expectations and remedies for this actor's 2024 campaign against water systems. (2) In practice, amateur pro-Russian hacktivists of the People's Cyber Army (Cyber Army of Russia Reborn) reached the HMI remotely, clicked through settings until valves opened, and wasted untreated water - impact confirmed by the city manager - and the utility detected none of it: the intrusion surfaced from the attackers' own Telegram video of 30 Aug 2024, was identified weeks later by an outside ICS researcher who reported it to CISA, and the city engaged the agencies only when they called it. The city's own response was minimisation - the city manager's public account was no big deal and an easy, quick fix, delegated to the external systems integrator. (3) This is a material-column knowing-doing gap: the remedies that would have blocked or revealed the intrusion - disconnecting the HMI from the public internet, removing default credentials, MFA, segmentation, logging remote HMI logins - were documented, published technical knowledge aimed precisely at small water utilities, yet none was realised in the utility's operations or architecture; the classification rests on the detection stage (clearest evidence: the victim's own account of learning about the intrusion from the attacker's video via external parties), not on the entry vector, which is not incident-confirmed and is not load-bearing for the type. CSO frames the facility as under-resourced rather than governed by policy that leadership deferred, so no leadership-level decision to ignore a doctrine is evidenced on the load-bearing stages; the response-stage minimisation (a Strategy to Leadership pattern) is recorded as a contributing gap only. RM27 (Expertise to Execution lens): the treatment HMI sat reachable from outside with no protective configuration standing between it and the internet, the exact exposure EPA/CISA document for these HMIs, in a small utility whose experts had neither the tooling nor the staff to maintain secure OT configurations (causal role: Enabling). RM34 (Expertise to Execution lens): no monitoring, alert handling or remote-login logging was in place to reveal unauthorised HMI control, so the city's time-to-detect was indefinite and the intrusion only surfaced via the attacker's public video and an outside researcher (causal role: Obscuring). RM08 (Expertise to Execution lens): the remedies were knowable but the utility demonstrably lacked the in-house capability to apply them - CSO presents it as one of the under-resourced facilities that are the lowest-hanging fruit of OT security, and the fix itself was assigned to an external systems integrator (causal role: Enabling). RM39 (Expertise to Execution lens): with no forensic capability of its own, the city could not scope or communicate the incident itself - disclosure was forced by the attacker's leak and press enquiries, and its public account stayed minimised (causal role: Constraining).

Evidence by stage

Risk markers

Sources

Notes

EuRepoC row; incident matches the stated victim (Stanton, Texas water system). Attribution is attacker self-attribution via a Telegram video of 30 Aug 2024 (EuRepoC attribution_basis: attacker confirms); the group's name varies across reporting (People's Cyber Army / Cyber Army of Russia (Reborn)), and a possible GRU Sandworm link is reported via Mandiant but unverified and not load-bearing. The entry vector for this specific intrusion is not disclosed; the VNC/default-password mechanism is the documented campaign pattern (CISA alert 25 Sep 2024, EPA/CISA fact sheet), so the entry stage is scored pattern-based. The attribution_source_url in the case file is a YouTube video, not opened as text here and not cited. The EPA Water Sector Incident Analysis on CARR (Oct 2024) referenced by WaterISAC sits behind member access and was not opened. Dataset field mitre_impact 'Firmware Corruption' does not match the public reporting (settings manipulation and wasted raw water, no firmware corruption) - data-quality remark. No pro-active public notification by the city to residents is documented before media reporting; no legal notification duty for a system this size is established in the sources opened, so RM39 rests on the attacker-forced disclosure and the minimised public account.