DedSec

PC repair and optimization, made simple.

DedSec is an open source tool by Mauricio Vuljevas for PC repair, diagnostics, resource optimization, safe cleanup, secure installer monitoring, and system maintenance.

The repository is a TypeScript monorepo with two product surfaces:

• Public website: product showcase, documentation, release notes, and download links.
• Desktop app: Electron application for Windows, macOS, and Linux.

Project metadata:

• Author: Mauricio Vuljevas
• Author URI: https://www.mvuljevas.com
• Repository: https://github.com/mvuljevas/DedSec

Current Status

The first application scaffold is active:

• apps/web builds a multilingual Next.js landing surface with /es and /en.
• apps/desktop builds an Electron/Vite/React shell with a typed preload boundary.
• packages/domain, packages/i18n, and packages/ui provide shared foundations.

Proposed Monorepo Layout

apps/
  web/       Public website, Next.js app.
  desktop/   Cross-platform Electron desktop app.
packages/
  domain/    Shared product types, validation schemas, and business rules.
  i18n/      Shared locale configuration and product messages.
  ui/        Shared UI primitives once web and desktop converge.
docs/
  adr/       Architecture decision records.

Recommended Stack

• Language: TypeScript.
• Package manager: npm workspaces.
• Website: Next.js with React.
• Desktop: Electron, React, Vite.
• Shared validation: Zod or similar schema library.
• Local desktop storage: SQLite when durable local state is required.
• Cloud backend: PostgreSQL-backed service when accounts, sync, or downloads require server state.

See docs/ARCHITECTURE.md for the full foundation analysis.

Development

Install dependencies:

npm install

Build everything:

npm run build

Type-check everything:

npm run typecheck

Run the website:

npm run dev:web

Run the desktop app:

npm run dev:desktop

Repository Workflow

DedSec follows the Gitflow and PR rules documented in AGENTS.md. All meaningful project changes should update docs/SNAPSHOTS.md and, when applicable, TECHDEBT.md.

For efficient assisted development, start with docs/AI_CONTEXT.md and use docs/AI_SEARCH.md before opening broad source trees.