GPU/CPU-Powered Rule Engineering & Performance Benchmarking Toolkit
A comprehensive suite of specialized tools for Hashcat rules processing — from intelligent deduplication and statistical analysis to GPU-accelerated ranking with memory-mapped file loading, performance benchmarking, and automated rule extraction.
Functional deduplication and semantic optimization of Hashcat rules using full rule engine simulation.
Statistical rule analysis and generation using Markov models and combinatorial validation.
GPU-accelerated rule performance evaluation with memory-mapped file loading and dual-metric scoring system.
Automated rule extraction from wordlist transformations using BFS-based GPU computation.
Remove invalid rules from Hashcat logs containing startup errors with client-side processing and auto-cleanup.
Generate Hashcat rules from email patterns with trailing digit extraction and domain filtering.
GPU-accelerated rule performance testing with advanced visualizations, statistical analysis, and configuration confirmation.
Complete Hashcat rules implementation for GPU processing with all major rule categories and optimizations.
| Tool | GPU Support | CPU Support | Performance Benchmarking | Max File Size | Rule Limit | Processing Type |
|---|---|---|---|---|---|---|
| Rule Minimizer | ✅ Optional | ✅ Multi-core | ❌ | Limited by RAM | Unlimited | Functional Deduplication |
| Concentrator | ✅ Combinatorial | ✅ Markov | ❌ | Limited by RAM | Unlimited | Statistical Analysis |
| Ranker | ✅ Full GPU | ❌ Limited | ✅ Basic | Limited by VRAM | 500,000+ | Performance Evaluation |
| Rulest | ✅ BFS Processing | ✅ Fallback | ❌ | Limited by VRAM | Unlimited | Rule Extraction |
| Rule Cleaner | ❌ Not Required | ✅ Client-side | ❌ | 100MB | Unlimited | Validation & Cleanup |
| Email Extractor | ❌ Not Required | ✅ Client-side | ❌ | 100MB | Unlimited | Pattern Recognition |
| Aether | ✅ Full GPU | ✅ Multi-core | ✅ Advanced | Limited by VRAM | Unlimited | Performance Testing |
| OpenCL Kernel | ✅ Full GPU | ❌ Not Applicable | ❌ | Limited by VRAM | Unlimited | Rule Processing Engine |
Functional Deduplication
Identifies and removes rules that produce identical transformations
Multi-Core Processing
Leverages all CPU cores for maximum performance
Levenshtein Filtering
Removes rules with similar functionality using edit distance
Markov Statistical Analysis
Analyzes rule patterns and generates statistically relevant new rules
Combinatorial Generation
Creates rule combinations based on frequency analysis
Multi-File Aggregation
Combines multiple rule files into a concentrated set
Memory-Mapped File Loading
10-50x faster file loading using mmap for direct file access
OpenCL GPU Acceleration
Massively parallel rule evaluation using GPU processing
Dual-Metric Scoring
Evaluates rules based on both performance and effectiveness
Fast Word Count Estimation
No more double file reading - estimates word count from file size
v3.2 Performance Features:
Before v3.2:
After v3.2:
BFS Rule Discovery
Breadth-first search for optimal rule chains
Leetspeak Support
Automatic detection and rule generation for leetspeak patterns
Multi-Depth Chaining
Finds complex transformation sequences
Client-Side Processing
All processing happens in your browser for privacy
Large File Support
Handles files up to 100MB efficiently
Auto-Cleanup
Automatic cleanup of processed files
Dual Format Support
Handles username+digits-tag@domain formats
Pattern Recognition
Identifies common email patterns and transformations
Domain Filtering
Focuses on specific domains or excludes others
OpenCL GPU Acceleration
Massively parallel rule processing using OpenCL for high-performance benchmarking
Advanced Visualizations
Radar charts, heatmaps, and statistical summaries for comprehensive analysis
Configuration Confirmation
Interactive setup with comprehensive configuration validation before execution
Built-in Test Words
50 carefully selected test words included - no external dictionaries needed
Complete Rule Implementation
All Hashcat rules implemented for GPU processing
GPU Optimization
Kernel code optimized for parallel execution
Ready-to-Use Code
Production-ready OpenCL kernel implementation