Free, native, privacy-first cleaner and application uninstaller for macOS.
Built with Swift 6 and SwiftUI. Moves files to Trash by default instead of permanently deleting them.
Requirements: macOS 26.0 or later. Apple Silicon only (Intel Macs are not supported).
✨ Features • 📸 Screenshots • 🛠️ Build from Source • 📖 Documentation
| Principle | Technical Implementation |
|---|---|
| No Telemetry | No background daemons, analytics SDKs, or remote loggers. All scanning runs locally without network access. |
| Trash-First | Files are moved to macOS Trash via trashItem(at:) when supported, preserving native file recovery. |
| Protected Paths | Hardcoded blocklists protect SIP paths, /System, /Library, /usr, ~/.ssh, and user document directories. |
| Open Core | Swift 6 codebase with strict concurrency, actors, and structured task groups. Core engine and heuristics are inspectable. |
| On-Device AI | Checksums, hashes, and FoundationModels file explanations execute strictly on-device. |
| Native Runtime | Built in Swift 6 and SwiftUI with Liquid Glass materials. No Electron or WebKit container overhead. |
View high-resolution screenshots
- Direct Download: Grab the latest
.dmgfrom GitHub Releases. - Mount & Drag: Open the
.dmgand dragMacOSCleaner.appinto/Applications. - First Launch (Gatekeeper): If macOS shows a Gatekeeper warning on locally built/unnotarized apps:
sudo xattr -r -c /Applications/MacOSCleaner.app
Note: This removes the quarantine attribute set by macOS. It does not modify app files, system security, or grant persistent permissions.
- Smart Cleanup: scans 54 categories and 1,770 rules across 251 applications, 66 developer toolchains (Xcode, Docker, Rust
target, Gopkg, Python.venv, Nodenode_modules), local AI models (Ollama, MLX, Hugging Face), and APFS purgeable space. - Forensic uninstaller inspects 30 evidence types (Bundle ID, Team ID, Spotlight metadata, Launch Services) to find remnants, with root-level helper removal and rollback snapshots.
- Duplicate Finder: identifies duplicate files through a 3-stage pipeline (file size matching, 4 KB header checksum, full SHA-256 verification).
- Disk space analyzer groups directory contents by file type (Videos, Audio, Photos, Documents, Archives) for inspection.
- Process and Service Manager: monitors live CPU and RAM usage with termination safeguards for critical processes (
kernel_task,launchd), plus LaunchAgents and LaunchDaemons toggling. - Native system integrations include on-device
FoundationModelsexplanations for unknown caches, Siri and App Intents automation, Liquid Glass keyboard navigation (⌘,,⌘C,⌘F,⌘R), and 10 language localizations.
Complete feature breakdowns and path specifications are documented in the MacOSCleaner Wiki.
Prerequisites: macOS 26.0+, Xcode 18+, XcodeGen.
git clone https://github.com/AlexTkDev/MacOSCleaner.git
cd MacOSCleaner/MacOSCleaner
xcodegen
open MacOSCleaner.xcodeprojBuild and run with ⌘R, or run tests with ⌘U.
Note: Public source builds use the public fallback cleanup catalog.
— Star the repository and watch releases for update notifications.
— Ask questions, propose features, and share feedback.
— Track upcoming tasks for the next release.
— Report bugs or suggest rule improvements.
— Architecture overview and developer guides.
— Explore repository online and open in Cursor.
Note: Contributions are subject to the project's Contributor License Agreement (CLA).
|
Aleksandr (AlexTkDev)
|
Citation reference:
CITATION.cff
Dual-licensed:
- GNU GPLv3 + Commons Clause v1.0 for personal, non-commercial use, testing, and open-source contributions. Commercial resale and commercial re-licensing are prohibited.
- Commercial License for enterprise, commercial, or proprietary distribution. Contact the author for licensing inquiries.
Trademark Notice: The license does not grant permission to use the project name ("MacOSCleaner"), logos, or branding in derivative works.