GOD-CWAK/README.md

GODCWAK

GPU-accelerated account-recovery lab gear with proxy rotation, targeted wordlist building, and local Ollama CAPTCHA tooling.

GODCWAK is a high-energy Windows desktop application for authorized recovery work, local security testing, and lab validation on systems you own or are explicitly permitted to test. It brings a cyberpunk CustomTkinter interface to password-candidate generation, proxy health checks, live attempt logging, and local AI-assisted CAPTCHA experiments.

This project is not a license to attack other people, harvest credentials, bypass access controls, or run unauthorized login attempts. Keep it legal, scoped, and documented.

What It Does Today

• Password Builder: creates targeted password candidates from known fragments, custom words, number ranges, symbols, capitalization rules, leet variants, pattern templates, and length/character constraints.
• GPU-aware generation: uses CuPy when a CUDA GPU is available, NumPy for accelerated CPU paths, and pure Python as the fallback.
• Proxy Manager: fetches public proxies, imports/pastes proxy lists, tests proxy health, removes dead entries, and commits valid proxies to the runtime pool.
• Attack Terminal: runs a configurable multi-worker login workflow against the selected target URL with proxy rotation, delay controls, timeout settings, live progress, pause/resume/stop controls, and attempt logging.
• Logs view: shows runtime activity and records attempts to data/attempt_log.csv.
• Setup tab: stores local Ollama settings, lists/tests models, and supports local model configuration for CAPTCHA-solving experiments.

Current Limits

• CAPTCHA detection is present in the attack workflow, but full CAPTCHA solve-and-submit automation is still roadmap work.
• Browser automation modules exist, but they are not yet connected into the attack flow.
• Login success classification and generic site handling are still under active improvement.
• Large generated wordlists and generated attempt logs can contain sensitive data; keep them local and out of commits.

See NEEDS_TO_BE_DONE.md for the bug list, audit notes, and roadmap.

Install On Windows

From PowerShell in the project directory:

py -3 -m venv .venv
.\.venv\Scripts\Activate.ps1
python -m pip install --upgrade pip
pip install -r requirements.txt

If PowerShell blocks activation scripts, run:

Set-ExecutionPolicy -Scope CurrentUser RemoteSigned

Then activate the virtual environment again.

Run

Double-click:

GODCWAK.bat

Or run from PowerShell:

.\GODCWAK.bat

The launcher prefers .venv\Scripts\python.exe, then the Windows py -3 launcher, then python from PATH.

You can also launch directly:

python main.py

Build Windows Release (PyInstaller)

Production builds produce a portable onedir folder under dist/GODCWAK/ and a zip under release/.

From PowerShell in the project directory:

# One-time: activate venv and install build deps
.\.venv\Scripts\Activate.ps1
pip install -r requirements-build.txt

# Build release (no console window)
python build\build_release.py --clean

# Troubleshooting build (shows console + logs)
python build\build_release.py --clean --debug

Or double-click build_release.bat at the project root.

Output layout

dist/GODCWAK/
  GODCWAK.exe              Main application
  Start GODCWAK.bat        Optional launcher
  config_ollama.json       Writable Ollama settings (seeded on first run)
  data/                    Passwords, proxies, attempt logs (created at runtime)
  _internal/               Bundled Python runtime and dependencies
release/
  GODCWAK-v1.0-win64-YYYYMMDD.zip

Notes

• Default release excludes CuPy/Numba/Playwright to keep the bundle smaller; use --gpu if CUDA libraries are installed and you want them bundled.
• Place assets/icon.ico before building to embed a custom application icon.
• Packaged builds log to data/godcwak.log and show error dialogs instead of a console window.

GPU Notes

GODCWAK checks for CuPy at startup. With cupy-cuda12x installed and a working NVIDIA CUDA-capable GPU, password-combination index generation can use the GPU for larger workloads. Without CUDA, the app falls back to NumPy CPU acceleration or pure Python.

If CuPy does not load, the app still runs. The startup log will report that GPU acceleration is disabled.

Ollama And CAPTCHA Notes

Ollama settings live in config_ollama.json. The default host is:

http://localhost:11434

Use the Setup tab to connect to a local Ollama server, list models, save a model choice, and run basic model tests. Vision-model CAPTCHA solving exists as a local solver module, but the full browser capture, solve, click/type, and retry loop is not yet wired into the attack workflow.

Project Structure

.
├── GODCWAK.bat              Windows launcher
├── main.py                  Application entry point
├── config.py                Paths, defaults, and runtime constants
├── config_ollama.json       Local Ollama and browser settings
├── requirements.txt         Python dependencies
├── NEEDS_TO_BE_DONE.md      Audit notes and roadmap
├── data/                    Local runtime data and logs
└── src/
    ├── attack/              Login client and multi-worker orchestrator
    ├── captcha/             Ollama solver and browser/screenshot helpers
    ├── engine/              Password generation and GPU acceleration
    ├── gui/                 CustomTkinter application tabs and theme
    ├── proxy/               Proxy fetch, validation, and pool management
    └── utils/               File and attempt logging helpers

Authorized Use Only

Use GODCWAK only for:

• Accounts and systems you own.
• Recovery workflows you are authorized to perform.
• Internal lab targets and controlled test environments.
• Defensive validation with written permission and a clear scope.

Do not use it for unauthorized credential attacks, account takeover, evasion, abuse, or activity that violates law, policy, or platform terms. The operator is responsible for scope, consent, logs, and consequences.

Roadmap

The punch list lives in NEEDS_TO_BE_DONE.md. Top priorities include wiring CAPTCHA automation into the attack flow, strengthening login classification, improving session/thread safety, refining proxy lifecycle handling, and adding automated tests.