Critical Infrastructure: Nuclear Power
AI doesn’t just analyze evidence—it decides how to analyze it. RemiFetch adapts forensic methodology in real time based on detected patterns, automatically applying the right techniques—correlation, artifact extraction, behavioral analysis, and timeline reconstruction—to match the investigation.
By linking signals across systems and evolving its approach as new evidence emerges, AI reveals relationships and attack paths that static workflows cannot. This enables deeper insight, faster investigations, and defensible, evidence-driven conclusions.
Supported Industrial Control Systems
- Siemens
- Rockwell Automation
- Emerson
- GE Vernova
- Schneider Electric
- ✓ Not limited to known malware signatures, static detections, or predefined attack rules
- ✓ Detects adversary behaviors, tradecraft patterns, and operational relationships across systems, accounts, and time
- ✓ Helps reconstruct coordinated intrusion activity associated with foreign adversaries and hostile cyber operations
- ✓ Identifies indicators that AI may have been leveraged for automated reconnaissance, access attempts, escalation, or disruption activity
- ✓ Correlates activity across devices, networks, platforms, vendors, and user accounts into a unified investigative view
- ✓ Exposes hidden attack paths, multi-stage operations, and coordinated actions that traditional tools often fail to connect
- ✓ Produces evidence-backed findings with defensible, audit-ready timelines for intelligence, investigative, and reporting use
Remi analyzes offline nuclear OT/IT event logs to surface insider behavior, disruption tradecraft, and coordinated activity

Remi identifies adversary tradecraft in nuclear environments by linking low-signal events into behaviors: access → staging → execution → concealment. It flags suspicious access patterns (privileged use, unusual endpoints, off-hours), operational disruption indicators (unexpected command/mode changes, safety-relevant state transitions, alarm floods/silence), and persistence/covering tracks (log gaps, time shifts, unexpected configuration changes). By correlating events across OT and supporting IT systems, Remi surfaces coordinated activity that looks benign in isolation but high-risk in sequence—then packages evidence and summaries for investigation and compliance.
Water & Wastewater (ICS) — Detection Catalog
Scrollable list (click a rule to expand)
Access & Authentication
Unauthorized HMI/EWS Login Access
Privileged Account Change Identity
Unapproved Remote Access Session Remote
Process & Command Integrity
Unauthorized Pump Start/Stop Control
Valve Actuation Outside Operating Window Change
Chemical Dosing Setpoint Change Setpoint
Rapid Command Oscillation (“Chatter”) Anomaly
Safety & Operations Monitoring
Tank Level / Flow Anomaly Spike Telemetry
Pressure/Backwash Cycle Anomaly Operations
Alarm Flooding / Alarm Silence Pattern Alarms
Asset & Configuration Changes
PLC/RTU Configuration Change Config
Firmware/Logic Change Detected Baseline
New OT Asset Discovered Inventory
Network & Segmentation
OT Zone Boundary Violation Network
New East–West Communication Path Network
Historian/Data Gateway Anomaly Telemetry
Data Integrity & Readiness
Time Sync Drift / Jump Time
Telemetry Gap / Dropout Integrity
Case Evidence Completeness Readiness
Nuclear Disruption Investigation Reports

Remi generates this nuclear-focused report set by reading your offline OT/IT event logs from the case workspace and reconstructing adversary (or insider) tradecraft as behavior, not isolated alerts. It looks for sequences consistent with disruption operations—access → positioning → execution → concealment—including credential/privilege misuse, unusual operator/engineering workstation activity, suspicious command origin patterns, abnormal mode/setpoint/control sequences, and safety-relevant state changes that require scrutiny. It also highlights concealment indicators such as atypical log gaps, time/clock anomalies, alarm flooding or silence patterns, and configuration/logic drift that may indicate tampering or unauthorized change pathways.
Report Types
Executive_Summary.txt
What it does: High-level case overview for leadership and fast triage—key findings, severity posture, and what requires immediate review.
Operational_Narrative.txt
What it does: Investigator narrative explaining suspected disruption tradecraft as behavior sequences, linking access, actions, context, and concealment indicators.
Reactor_Event_Timeline.txt
What it does: Chronological timeline of significant events with timestamps to reconstruct “what happened when.”
Safety_System_Activity.txt
What it does: Focused view of safety-relevant activity—critical-state context and transitions requiring heightened scrutiny.
Control_Actions.txt
What it does: Summary of control actions (commands/mode changes) grouped for review of origin consistency, intent, and potential impact.
Critical_Event_Log.txt
What it does: Prioritized top-risk events for rapid triage—high-signal anomalies and safety-impacting indicators surfaced ahead of noise.
Artifact_Inventory.txt
What it does: Inventory of collected/produced artifacts and outputs to support defensible review and export packaging.
Artifact_Origin_Report.txt
What it does: Provenance and traceability—maps artifacts back to sources and acquisition context for auditability.
Evidence_Excerpts.txt
What it does: Curated excerpts of key evidence—short, relevant slices with enough context to support investigation and reporting.
Thread_Index.txt
What it does: Index of correlated investigative threads—links related events across users/hosts/assets/time windows to follow multi-step tradecraft.