SentraCore is a local system behavior intelligence platform designed to continuously analyze telemetry, understand performance behavior, detect anomalies, explain slowdowns, and estimate future system pressure before it impacts responsiveness.
Rather than focusing on isolated resource snapshots, SentraCore interprets system activity over time to answer three critical questions:
What is happening?
Why is it happening?
What is likely to happen next?
Built for intelligent desktop observability and explainable operational insight.
SentraCore combines telemetry collection, behavioral modeling, anomaly detection, predictive analytics, and explainable diagnostics into a unified local intelligence system.
The platform continuously learns machine-specific operating patterns and transforms raw system measurements into actionable insights.
SentraCore currently includes:
- Real-time telemetry monitoring
- Adaptive behavioral baseline learning
- Statistical anomaly detection
- Root cause analysis
- Predictive risk estimation
- Historical monitoring and logbook tracking
- Flutter desktop dashboard
- Desktop packaging and installer workflow
Unified system health scoring generated from:
- Resource pressure
- Behavioral deviation
- Sustained stress trends
- Predictive degradation indicators
Learns normal operating behavior per machine, including:
- CPU utilization patterns
- Memory behavior
- Disk activity trends
- Time-based workload characteristics
The objective is to detect meaningful deviation rather than isolated spikes.
Correlates multiple signals to identify likely contributors to degradation.
Analysis includes:
- Process activity
- Resource contention
- Event timing
- Performance degradation patterns
Outputs ranked probable causes instead of raw metric streams.
Forecasts future pressure using historical and trend-based analysis.
Includes:
- Memory saturation estimation
- CPU trend projection
- Disk pressure forecasting
- Estimated degradation warnings
Continuously records and visualizes:
- CPU pressure
- Memory pressure
- Disk pressure
- Long-term behavioral trends
- Historical intelligence events
Includes:
- Live alerts
- Alert history
- Diagnostics timeline
- Windows notifications
- Root cause summaries
Flutter Dashboard
↕
WebSocket / REST API
↕
Python Intelligence Engine
│
├── Telemetry Collection
├── Baseline Learning
├── Anomaly Detection
├── Correlation Engine
├── Prediction Engine
└── Alert System
| Layer | Technology |
|---|---|
| Engine | Python 3.11 · psutil |
| API | FastAPI · WebSockets |
| Dashboard | Flutter Desktop |
| Packaging | PyInstaller · Inno Setup |
| Automation | GitHub Actions |
See project documentation for full setup instructions.
Start the engine:
.venv\Scripts\python -m engine.mainLaunch dashboard:
cd dashboard
flutter run -d windowsDownload the latest desktop release from:
GitHub Releases
Installer capabilities:
- Installs SentraCore
- Creates application shortcuts
- Configures optional startup behavior
- Deploys the monitoring engine
- Prepares runtime dependencies
| Document | Description |
|---|---|
docs/setup/development_setup.md |
Development environment setup |
docs/setup/engine_setup.md |
Engine installation and configuration |
docs/setup/dashboard_setup.md |
Dashboard build and execution |
docs/architecture/intelligence_layer.md |
Intelligence pipeline internals |
docs/architecture/building.md |
Packaging and release workflow |
SentraCore is built around six guiding principles:
- Observation
- Behavioral Modeling
- Pattern Recognition
- Correlation Analysis
- Predictive Awareness
- Explainable Intelligence
| Requirement | Version |
|---|---|
| Python | 3.11+ |
| Flutter | Stable 3.x+ |
| Platform | Windows (Primary) |
Linux and macOS currently support development workflows but Windows remains the primary runtime and packaging target.
engine/
dashboard/
docs/
installer/
pages/
assets/
tests/
Licensed under the Apache License 2.0.
See: