Unified Hashcat Rule Processor — Extract, generate, and process hashcat password rules with GPU acceleration, Markov chain modeling, and functional minimization.

• OpenCL GPU Acceleration — Batch rule validation offloaded to GPU for high throughput
• Three Processing Modes — Extraction, Combinatorial generation, and Markov-based generation
• Hashcat Rule Engine Simulation — Full CPU-side implementation of hashcat's rule operators
• Functional Minimization — Deduplicate rules that produce identical outputs
• Levenshtein Distance Filtering — Remove near-duplicate rules by edit distance
• Memory Safety — Monitors RAM/swap usage with configurable thresholds and disk-spill mode
• Multiple Output Formats — line (compact) or expanded (operator + args separated by spaces)
• Interactive & CLI Modes — Guided wizard or full argument-driven usage

Python 3.8+

sys, os, re, argparse, math, itertools, multiprocessing, tempfile, random, datetime, threading, collections, typing

All optional packages degrade gracefully — the tool runs on pure Python if none are installed.

git clone https://github.com/youruser/concentrator.git cd concentrator pip install pyopencl numpy tqdm psutil # optional but recommended

Run without arguments to launch the guided wizard:

python concentrator.py

python concentrator.py [OPTIONS] FILE_OR_DIRECTORY [FILE_OR_DIRECTORY ...] 

One mode flag is required.

Extract the most frequent (or statistically weighted) rules from existing rule files.

# Extract top 5000 rules by frequency python concentrator.py -e -t 5000 rules/ # Extract top 10000 rules sorted by Markov sequence probability python concentrator.py -e -t 10000 -s rules/*.rule

Generate rules by exhaustively combining the most common operators up to a target count.

# Generate 50k rules using operator combinations of length 2–4 python concentrator.py -g -n 50000 -l 2 4 hashcat/rules/

Generate statistically probable rules using a Markov chain model trained on the input rule files.

# Generate 10k Markov rules of length 1–5 python concentrator.py -gm -gt 10000 -ml 1 5 hashcat/rules/

Load, validate, deduplicate, and functionally minimize existing rule sets interactively.

# Process rules using disk mode to avoid RAM exhaustion python concentrator.py -p -d rules/ # Process with Levenshtein distance filtering (max dist 3) python concentrator.py -p -ld 3 rules/

line — Standard hashcat rule format, one rule per line:

i3li4ei5y li6po7io8e D2i41o50 D1$9$6$0 o6g$1$5 D1i3hi4o 

expanded — Each operator and its arguments separated by spaces, one rule per line:

i3l i4e i5y l i6p o7i o8e D2 i41 o50 D1 $9 $6 $0 o6g $1 $5 D1 i3h i4o 

Concentrator recursively scans directories up to 3 levels deep for files with these extensions:

.rule .rules .hr .hashcat .txt .lst

You can pass individual files, directories, or a mix of both.

Concentrator validates and simulates the full hashcat rule operator set, including:

# Extract top 5000 rules (GPU off) from a glob python concentrator.py -e -t 5000 --no-gpu rules/*.rule # Generate 100k combinatorial rules, output in expanded format python concentrator.py -g -n 100000 -l 1 3 -f expanded hashcat/rules/ -ob my_rules # Markov generation with custom length range python concentrator.py -gm -gt 20000 -ml 2 4 hashcat/rules/ # Process and minimize rules, writing temp files to /tmp/scratch python concentrator.py -p -d --temp-dir /tmp/scratch rules/ # Interactive mode python concentrator.py

• At startup, Concentrator prints current RAM and swap usage.
• If RAM usage exceeds 85%, a warning is raised and you are prompted before continuing.
• Use --use-disk / -d with -p mode to spill intermediate data to disk.
• Use --in-memory to force full in-RAM processing (fastest, but watch your available memory).
• Install psutil to enable memory monitoring.

MIT License. See LICENSE for details.

• Penguinkeeper for testing rules
• https://github.com/0xVavaldi/ruleprocessorY
• https://github.com/synacktiv/rulesfinder
• https://github.com/mkb2091/PyRuleEngine/blob/master/PyRuleEngine.py
• https://github.com/hashcat/hashcat-utils/blob/master/src/cleanup-rules.c