JARVIS: AI-Enabled Home & Desktop Assistant

JARVIS is a powerful, low-latency, and beautifully designed AI-enabled home and desktop assistant. Built as a desktop application using Tauri v2, React, TypeScript, and Rust, JARVIS is engineered for performance, privacy, and extensibility.

---

📌 Implementation Status

🟢 Local Mode (Fully Active)

• Encrypted Offline Protocol: Full local execution prioritizing privacy.
• AI & MCP Integrations: High-performance local AI coordination using agent_rs and rig-core.
• Parakeet Transcription: Real-time voice listening and local speech-to-text processing using the Parakeet model.
• Telemetry HUD: Real-time monitoring of local CPU, RAM, and network statistics using Rust's sysinfo and rendered via interactive, fluid Recharts components.
• Session Management: Full persistence of chat histories and sessions stored locally in a SQLite database.

🟡 Sync Mode (Future Roadmap)

• Cloud & Fleet Synchronicity: Coordinate and monitor multiple linked nodes.
• Device Management & WOL: Remotely link new devices and wake them via Wake on LAN (WOL).
• Distributed Automation: Create, edit, and orchestrate automation routines across your local node fleet.

---

📸 Interface & Feature Gallery

1. Boot & Mode Protocol Selection

When JARVIS boots, users are prompted with a secure, styled protocol selection overlay to activate either Local Mode or Sync Mode.

Startup Interface	Protocol Selection

Local Mode (Offline Conversation Interface)	Sync Mode (Online Dashboard Preview)

---

2. Main Dashboard & Local Telemetry HUD

Once inside Local Mode, users have a split terminal-style chat interface on the left and a collapsible Telemetry HUD on the right, providing live performance metrics.

Collapsible HUD Overview	Custom Metrics Graphs

Tip: The HUD can be collapsed or rearranged dynamically for a tailored developer workspace.

---

3. AI Copilot Chat & Session Management

JARVIS includes a robust session manager allowing users to rename, create, or delete separate local conversations. File attachment support and MCP tool integrations allow rich commands.

Chat with MCP Tools	File Attachment

Session List Manager	System Configuration

---

🛠️ Technology Stack

Frontend Architecture

• Framework: React 19 (Vite + TypeScript)
• Styling: Tailwind CSS v4 (Theme variables configured inside the @theme directive in src/styles.css, supporting dynamic, interchangeable skins like theme-amber, theme-blue, and theme-red).
• Animations: Framer Motion for premium micro-animations, fade-ins, and page transitions.
• Visualization: Recharts for live hardware graphs.
• State Management: Zustand stores.

Backend Architecture (Rust)

• Framework: Tauri v2
• Database: SQLite (managed locally via rusqlite with bundled features).
• Transcription: Local speech-to-text powered by the jarvis-transcriber library and the Parakeet offline model.
• Telemetry: System diagnostics using the Rust sysinfo library.
• Agents: Integration with LLM frameworks via agent_rs and rig-core.

---

📂 Repository Structure

Frontend (src/)

src/
├── assets/          # JARVIS branding, icons, and UI sounds
├── components/      # Global, stateless/dumb UI elements (Gauges, Cards, Toggles)
├── features/        # Feature-based domain logic (MCP integrations, node fleet management)
├── hooks/           # useWebSocket, useHardwareStatus, useSystemData
├── layouts/         # DashboardLayout (Titlebar, Sidebar, Notifications)
├── lib/             # Third-party configuration profiles
├── pages/           # Thin route views (DashboardPage, ModeSelectionPage, OfflineDashboardPage)
├── services/        # Bridge layer calling Tauri commands and REST APIs
├── store/           # Global Zustand states (Active Nodes, AI Context)
├── styles.css       # Core Tailwind CSS v4 design tokens and layouts
└── types/           # TS Interfaces for Nodes, Devices, and MCP schemas

Backend (src-tauri/)

src-tauri/
├── src/
│   ├── commands/    # Tauri command controller handlers (chat, config, system, voice)
│   ├── domain/      # Configuration, database models, voice and system structs
│   ├── handlers/    # Background daemon workers and transcription loops
│   ├── infrastructure/ # System telemetry collectors and SQLite DB manager
│   ├── lib.rs       # App setup, state management, and command registration
│   └── main.rs      # Tauri boot entry point
└── tests/           # Integration and unit tests

---

🚀 Getting Started

Prerequisites

• Rust (latest stable toolchain)
• Bun (recommended package manager) or Node.js

1. Clone & Install Dependencies

git clone https://github.com/skaarfundgandr/JARVIS.git
cd JARVIS
bun install

2. Run in Development Mode

To launch the desktop application:

bun run tauri dev

To run only the Vite frontend dev server (default port 1420):

bun run dev

3. Build for Production

To bundle the desktop executable for your operating system:

bun run tauri build

---

💻 Developer Command Reference

Rust Development

All Rust commands should be executed from within the src-tauri/ directory.

# Check Rust compilation
cargo check

# Run backend unit and integration tests
cargo test

# Run Rust linter (Clippy)
cargo clippy --all-targets --all-features -- -D warnings

# Run Rust code formatter
cargo fmt --all

---

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.