This repository has been archived as of 2026-09-07 (v6.0 "Autumn").
Meteorological Autumn 2026 begins on 2026-09-01 and the astronomical Autumnal Equinox in 2026 begins on 2026-09-22. This release is named "Autumn" to reflect the 2026 Autumn timeframe.
References:
This repository is archived and will no longer receive feature updates or upgrades. The project release v6.0 "Autumn" is the final release. The program can no longer be upgraded from this point — there will be no further upgrade path provided by the maintainers.
If you need to use the software, treat this repository as read-only. Take care to only run the code in controlled test environments (virtual machines or disposable systems) and follow the safety guidance in the README (dry-run mode and explicit enablement required for destructive actions).
# BlueScreen Trigger v5.0
Enterprise-grade Windows BSOD (Blue Screen of Death) generator with machine learning analytics, predictive modeling, and advanced system monitoring.
## Overview
**BlueScreen Trigger** is a sophisticated tool for triggering Windows system crashes with progressive versions offering increasingly advanced features:
- **v1**: Basic BSOD trigger with error handling
- **v2**: Async support, enhanced logging, CLI arguments
- **v3**: Recovery mechanisms, system snapshots, multiple trigger methods
- **v4**: Database persistence, REST API, scheduling, telemetry, plugins
- **v5**: Machine learning, anomaly detection, predictive analytics, visualization
## Features
### Core Functionality
- Trigger Windows BSOD via NtRaiseHardError Windows API call
- Multiple error code options (INVALID_IMAGE_FORMAT, FATAL_USER_CALLBACK_EXCEPTION, etc.)
- Dry-run mode for safe testing without actual system effects
- Comprehensive logging and error handling
- Platform verification and privilege escalation
### Version 5 Enhancements
- **Machine Learning Models**
- Neural network-based success prediction
- Isolation Forest anomaly detection
- Model persistence and training
- **Metrics Collection**
- Real-time CPU, memory, disk, and network monitoring
- Historical metrics storage in SQLite
- Automatic baseline collection
- **Analytics Engine**
- 24-hour trend analysis
- Optimal trigger time prediction
- Statistical calculations (mean, std, min, max)
- System load forecasting
- **Anomaly Detection**
- Detect unusual system behavior patterns
- Real-time anomaly scoring
- Configurable thresholds
- **Predictive Analytics**
- Forecast trigger success probability
- Recommend optimal system parameters
- Confidence scoring
- **Data Visualization**
- Interactive timeline charts (Plotly)
- Anomaly detection visualizations
- Comprehensive analytics dashboards
- Beautiful HTML reports
## Installation
### Requirements
- Windows OS (7+)
- Python 3.8+
- Administrator privileges (recommended)
### Setup
1. Clone the repository:
```bash
git clone https://github.com/arnikipad/bluescreen.git
cd bluescreen- Install dependencies:
pip install -r requirements.txt- (Optional) Install as package:
pip install -e .Collect system metrics:
python bluescreen_v5.py --dry-run --collectTrain ML models on historical data:
python bluescreen_v5.py --train --hours 24Analyze current system metrics:
python bluescreen_v5.py --analyzePredict trigger outcome:
python bluescreen_v5.py --predictGenerate visualizations:
python bluescreen_v5.py --visualizeGenerate analytics report:
python bluescreen_v5.py --reportShow system summary:
python bluescreen_v5.py --summaryExecute trigger (requires no --dry-run):
python bluescreen_v5.py --trigger--dry-run # Run without actual system effects
--hours N # Specify hours of historical data to use
--collect # Collect system metrics
--train # Train ML models
--analyze # Perform metrics analysis
--predict # Generate trigger predictions
--visualize # Create visualization dashboards
--report # Generate comprehensive report
--summary # Show system summary
--trigger # Execute the triggerREST API server:
python bluescreen_v4.py --apiScheduled trigger:
python bluescreen_v4.py --schedulerCheck status:
python bluescreen_v4.py --statusView history:
python bluescreen_v4.py --history 50With recovery points:
python bluescreen_v3.py --dry-run --method ntraiseWith event logging:
python bluescreen_v3.py --event-log --log-file system.logWith delay:
python bluescreen_v2.py --delay 5 --error-code PRIVILEGED_INSTRUCTIONBasic execution:
python bluescreen_v1.pyBlueScreenTriggerV5
├── MetricsCollector
│ └── SystemMetrics
├── AnomalyDetector
│ ├── Isolation Forest Model
│ └── StandardScaler
├── TriggerSuccessPredictor
│ ├── Neural Network (MLP)
│ └── StandardScaler
├── AnalyticsEngine
│ └── Trend Analysis
├── VisualizationEngine
│ ├── Plotly Visualizations
│ └── HTML Dashboards
└── BlueScreenLogger
└── Multi-Handler Logging
Metrics Table:
- timestamp, cpu_percent, memory_percent, disk_percent
- process_count, network stats, disk I/O
- context switches, interrupts
AnomalyDetector:
- Algorithm: Isolation Forest
- Features: 10 system metrics
- Contamination: 10% (configurable)
- Output: Anomaly score [0.0, 1.0]
TriggerSuccessPredictor:
- Algorithm: Multi-Layer Perceptron (Neural Network)
- Hidden Layers: 128 → 64 → 32 neurons
- Activation: ReLU
- Output: Success probability [0.0, 1.0]
general:
dry_run: true
obfuscate: false
log_file: bluescreen_v4.log
database:
enabled: true
path: bluescreen_v4.db
api:
enabled: false
host: 127.0.0.1
port: 8888
auth_enabled: false
schedule:
enabled: false
interval_seconds: 3600
telemetry:
enabled: false
endpoint: ""
trigger:
method: ntraise
error_code: INVALID_IMAGE_FORMAT
timeout_seconds: 10===== BLUESCREEN V5 ANALYTICS REPORT =====
Generated: 2026-09-05T13:39:45Z
SYSTEM METRICS:
CPU Usage: 45.2%
Memory Usage: 62.8%
Disk Usage: 78.1%
Process Count: 245
TRENDS (24h):
CPU: stable (mean: 42.3%, std: 8.5%)
Memory: increasing (mean: 58.2%, std: 12.1%)
Disk: stable (mean: 75.9%, std: 2.3%)
PREDICTION:
Success Probability: 0.89
Anomaly Score: 0.123
Recommended Delay: 2.5s
Confidence: 0.85
========================================
metrics_timeline.html- Interactive CPU/Memory timelineanomaly_detection.html- Anomaly scores over timedashboard.html- Comprehensive analytics dashboard
GET /api/healthGET /api/historyGET /api/statsGET /api/pluginsPOST /api/trigger
Authorization: Bearer <token>bluescreen/
├── bluescreen_v1.py # Basic version
├── bluescreen_v2.py # Async version
├── bluescreen_v3.py # Recovery version
├── bluescreen_v4.py # Enterprise version
├── bluescreen_v5.py # ML version
├── setup.py # Package setup
├── requirements.txt # Dependencies
├── bluescreen_config.yaml # Configuration
└── README.md # This file
To add custom ML models to v5:
from bluescreen_v5 import BlueScreenTriggerV5
trigger = BlueScreenTriggerV5()
# Train models
trigger.train_models(hours=24)
# Make predictions
metrics = trigger.metrics_collector.collect_metrics()
prediction = trigger.predict_trigger_outcome(metrics)
print(prediction.success_probability)- Install PyInstaller:
pip install pyinstaller- Build single-file executable:
pyinstaller --onefile --windowed bluescreen_v5.py- Build with all dependencies:
pyinstaller --onefile --hidden-import=sklearn --hidden-import=plotly bluescreen_v5.py- Executable will be in
dist/directory
pip install cx_Freeze
cxfreeze bluescreen_v5.py --target-dir dist- Collection: ~200ms
- Storage: ~50ms
- Total: ~250ms per cycle
- Anomaly Detector: ~2-5 seconds (100+ samples)
- Success Predictor: ~3-8 seconds (100+ samples)
- Combined: ~5-13 seconds
- Per metric: ~10-20ms
- Batch (100): ~200-400ms
Supported NTSTATUS error codes:
0xC000007B- INVALID_IMAGE_FORMAT (default)0xC000013B- FATAL_USER_CALLBACK_EXCEPTION0xC0000096- PRIVILEGED_INSTRUCTION0x00000050- PAGE_FAULT_IN_NONPAGED_AREA
- Ensure at least 20 historical metrics are collected
- Check that metrics database exists and has data
- Verify sklearn version is 1.0+
- Install plotly:
pip install plotly - Check that matplotlib is installed for fallback
- Ensure dashboards/ directory is writable
- Verify port 8888 is not in use
- Check firewall settings
- Ensure admin privileges if needed
- Run with administrator privileges
- Verify Windows platform (not WSL)
- Check that ntdll.dll is accessible
- Review log files for detailed errors
- Testing environments
- Virtual machines
- Systems with full backups
- Development/research contexts
- Model files are stored unencrypted
- Metrics database contains performance data
- API tokens should use strong authentication
- Enable encryption at rest for production
This project is provided for educational and authorized testing purposes only.
By using this tool, you accept full responsibility for:
- System damage or data loss
- Unauthorized system access or modification
- Compliance with applicable laws and regulations
- Authorized use only in environments you own or have permission to test
This tool should only be used by authorized administrators on systems where you have explicit permission.
For issues, questions, or contributions:
- Open an issue on GitHub
- Review the detailed logs in bluescreen_v*.log
- Check the configuration in bluescreen_config.yaml
- Consult the inline code documentation
BlueScreen Trigger v5.0 - Enterprise BSOD Generator with ML Analytics