WaTauri

Overview

WaTauri is an experimental, local-first desktop Whatsapp Client built with Tauri.

It connects to WhatsApp Web, syncs chats and messages into a local SQLite database, and exposes them through a desktop UI. The project is open source and currently under active development.

Why this exists

I have an ick for the official WhatsApp Web experience.

I'm basically trying to explore a desktop-first experience, with stronger organization, more power user features like, better search, better bulk actions, improved media workflows, and more control in general.

Obviously, this should do everything that WhatsApp Web currently does (yet to be achieved), but faster and more reliable and then some.

Status

This project is early-stage and experimental.

Core chat/message syncing is being built out, but the app is not ready yet.

Features

Look at FEATURES.md for everything.

• WhatsApp QR pairing
• Local SQLite message storage
• Chat list from local data
• Message history view
• History sync persistence
• Contact storage basics
• Sending messages
• Message pagination
• Read receipts
• Real-time updates with Server-Sent Events
• Contact name sync
• Group metadata
• Group participants (in progress; joined-group sync is implemented)
• Contact and group avatars
• Communities and related group metadata
• Pinned chats
• Pinned messages
• Favourite chats and contacts
• Archive workflows
• Safer local API routing
• Custom chat groups and filters
• Fast full-text search
• Bulk message/media actions
• Better media download/export
• Voice message transcription
• Desktop notifications
• Keyboard shortcuts
• Message scheduling
• Multiple accounts
• Privacy controls for typing/read signals

Screenshots

None yet.

Tech Stack

Desktop: Tauri
Frontend: Next.js, React
Backend: Go
Database: SQLite
Whatsapp Client: whatsmeow
Package manager: bun

Requirements

• Go
• Bun
• Rust and Cargo
• Tauri CLI dependencies

Getting started

Clone the repo

git clone https://github.com/your-username/whatsapp-tauri.git
cd whatsapp-tauri/whatsapp-tauri

Install the dependencies:

bun install

Run the app in development:

bun run tauri dev

Other useful scripts: bun run build (typecheck + build frontend), bun run preview (preview the production build).

Compiling the backend seperately:

cd src-go && go build -o backend .

This has been changed to now run automatically when tauri is called, however it does not recompile on every save.

Usage

1. Start the app with bun run tauri dev.
2. Open the desktop window.
3. Pair your whatsapp account using the QR Code.
4. Voila, (it'll take some time to load all the messages)

Runtime data is stored locally using SQLite.

Development

This project has three main parts:

• Next.js frontend
• Tauri Desktopo shell
• Go backend using whatsmeow

More info will be added later.

Project Structure

whatsapp-tauri/
  app/                 Next.js frontend
  public/              Static assets
  src-go/              Go backend and WhatsApp client
  src-tauri/           Tauri desktop shell
  docs/                Project notes and API docs

Roadmap

Phase 1: Reliable Core Client

• WhatsApp pairing
• Local message persistence
• Historical chat sync
• Sending messages
• Contacts, group metadata, participants, and avatars
• Message pagination
• Real-time updates with Server-Sent Events
• Read receipts

Phase 2: Better Desktop Client

• Desktop notifications
• Keyboard shortcuts
• Fast startup from local cache
• Better window/session restore
• Attachment handling
• Media previews

Phase 3: Power-User Organization

• Custom chat groups
• Saved filters
• Per-group notification rules
• Archive workflows
• Pinned chats
• Pinned messages
• Favourite chats and contacts
• Starred message workflows
• Multi-select chats/messages

Phase 4: Search and Export

• Full-text message search
• Search by sender/date/media type
• Voice note transcription
• Bulk media download
• Chat/message export

Phase 5: Experimental Features

• Scheduled messages
• Multiple accounts
• Privacy controls
• One-time media UX

Security

This app runs a local backend and stores WhatsApp session data locally. Do not commit: *.db *.db-wal *.db-shm wa-session.db userdata.db The backend should only be exposed locally. Do not run it on a public interface unless you understand the security implications.

Contributing

Contributions are welcome. I know that it's not properly organized yet, but I plan on using issues with a proper roadmap later on after the initial stage of development.

Disclaimer.

There is a fair bit of AI generated code and documentation. This project is not affiliated with WhatsApp, Meta, or any official WhatsApp product. It uses WhatsApp Web behavior through whatsmeow. Use at your own risk.