Windows Deep Cleaner

A Windows-first desktop cleaner project focused on safe deep cleaning, four-category classification, recoverable cleanup, and optional AI enhancement.

Current status

• Frontend desktop shell is scaffolded with React, TypeScript, Vite, and Vitest.
• Home, scan, results, and recovery screens are implemented as a working prototype.
• Frontend tests pass.
• Frontend production build passes.
• Rust/Tauri backend has not been committed yet because the local MSVC + Windows SDK environment is still not fully usable for linking.

Commands

npm install
npm test
npm run build

Windows native toolchain

Rust and future Tauri commands should be run through the Visual Studio developer environment.

Examples:

powershell -ExecutionPolicy Bypass -File .\scripts\with-msvc-env.ps1 "cargo test --manifest-path src-tauri\Cargo.toml --test domain_models"
powershell -ExecutionPolicy Bypass -File .\scripts\with-msvc-env.ps1 "cargo build --manifest-path src-tauri\Cargo.toml"

Docs

• Design spec: docs/superpowers/specs/2026-04-19-windows-cleaner-design.md
• Implementation plan: docs/superpowers/plans/2026-04-19-windows-deep-cleaner-v1.md

AI enhancement boundary

AI providers are optional. They can explain unfamiliar files and suggest better wording for borderline items.

AI providers cannot:

• override protected-item rules
• bypass user confirmation
• execute deletion directly