DiskPilot

A Windows disk cleanup and organization tool. Scans all drives, intelligently categorizes files into confidence tiers, and lets you clean up junk with a single click.

Features

• Smart scanning - Walks all connected drives and catalogs every file into a SQLite database
• 4-tier confidence system - Files are sorted into SAFE / RECOMMENDED / SUGGESTED / ASK based on what they are and how stale they are
• Developer-friendly - Automatically skips node_modules, .venv, __pycache__, .gradle, .cargo, and 30+ other dev directories
• Piracy detection - Flags known cracks, keygens, and activators as security risks
• Duplicate finder - Detects duplicate files across drives and suggests keeping the newest copy
• Version detection - Finds older versions of the same installer and flags them
• Drive organization - Assign purposes to drives (Games, Projects, Media, etc.) and find misplaced files
• 3-layer system protection - System files are excluded at scan, analysis, and deletion layers
• Recycle bin support - Defaults to recycle bin so nothing is permanently lost
• Dark modern UI - Built with CustomTkinter + ttk.Treeview for a responsive dark interface

Screenshot

Installation

From installer (recommended)

Download DiskPilot_Setup.exe from Releases and run it.

From source

git clone https://github.com/tehgee42/DiskPilot.git
cd DiskPilot
pip install customtkinter pillow
python app.py

Build the .exe yourself

pip install pyinstaller
python build.py
# Output: dist/DiskPilot.exe

Architecture

app.py              - Main window, ttk dark theme setup
panel_center.py     - Full UI layout (top bar, tier cards, filters, file list, bottom bar)
widgets.py          - FileListWidget (ttk.Treeview), TierCard, ToastWidget
panel_right.py      - Organize drives dialog (assign purposes, find misplaced files)
scanner.py          - File system walker, categorizer, system protection
smart_analyzer.py   - 4-tier confidence engine, duplicate/version/piracy detection
analyzer.py         - Aggregate queries, format helpers
organizer.py        - Delete queue, recycle bin support, drive purpose templates
database.py         - SQLite with WAL mode, batch inserts
theme.py            - Color palette, font constants
build.py            - PyInstaller build script
installer.iss       - Inno Setup installer script

How it works

1. Scan - Walks every drive (skipping system dirs and dev folders), catalogs file metadata into SQLite
2. Analyze - Queries the database to find junk, old installers, duplicates, stale files, and piracy artifacts
3. Tier - Each file gets a confidence tier and a human-readable reason explaining why it was flagged
4. Clean - Select files (or use quick-select buttons), hit Clean, files go to recycle bin

Confidence tiers

Tier	What it means	Examples
SAFE	Definitely deletable	Temp files, cache, logs
RECOMMENDED	Very likely safe	Old installers, torrents, duplicates
SUGGESTED	Probably fine, worth a glance	Stale archives, old backups
ASK	Only you know	Old media, abandoned projects

Developer directory skip list

The scanner automatically skips these directories to avoid indexing dev dependencies:

node_modules, __pycache__, .venv, venv, .env, .next, .nuxt, .cache, .npm, .yarn, .gradle, .cargo, .m2, .tox, .pytest_cache, .mypy_cache, .ruff_cache, .parcel-cache, .turbo, .angular, .svelte-kit, .expo, bower_components, .sass-cache, coverage, .nyc_output, .idea, .vs, vendor, target

System protection

Files are protected at three layers:

1. Scanner - System files (pagefile.sys, ntuser.dat, etc.) are categorized as system_protected and system directories (Windows, Program Files, etc.) are skipped entirely
2. Analyzer - All queries exclude system_protected category
3. Organizer - Delete execution refuses to touch any path containing system path fragments

Requirements

• Windows 10/11
• Python 3.10+ (if running from source)
• Database size scales with total files across all drives (e.g. ~50 MB for a small setup, ~2 GB for 9+ TB of storage)

Donate

If you find DiskPilot useful, consider supporting development:

License

MIT License. See LICENSE for details.