🤖 Omni-Bot: Multi-Modal AI Companion

Omni-Bot is a comprehensive AI companion platform featuring modular, hardware-first AI agents designed for edge devices.

📂 Projects

Jarvis-Bud 🤖

Type: Raspberry Pi Voice Assistant | Status: Active Development

A modular, voice-activated AI Agent for Raspberry Pi Zero 2 W with Whisplay HAT and PiSugar 3.

Features:

✅ Interactive First-Boot Wizard (Hardware > Connectivity > Personality > API)
✅ 4 Pre-built "Bud" personalities (Tech-Support, Trail-Runner, Hacker, Zen)
✅ Dual-Route Intelligence (Local Ollama + Cloud OpenRouter)
✅ Terminal-Dark UI with neon accents & 60fps animations
✅ Real-time waveform visualization during voice capture
✅ Battery management with PiSugar 3 monitoring
✅ Physical GPIO buttons for Wake/Control
✅ Fully async event loop for responsive UI

Quick Start:

cd Omni-Bot
bash setup.sh
python3 -m jarvis_bud.main

📖 Full Jarvis-Bud Documentation →

⚡ Final Deployment Mode (Pi Zero 2 W)

This branch ships a hardened offline deployment stack with:

🔐 Tamper-evident audit chain (logs/audit_chain.jsonl)
🛑 Long-press Button B emergency stop (kills tracked process groups)
⚠️ High-risk action gating (risk >= 6 requires explicit button confirmation)
🧠 Offline brain planner + reflection loop with GGUF model fallback chain
🧵 Tiny memory layer (FAISS/MiniLM optional, JSONL fallback always available)
📦 python main.py export for single-file HTML + encrypted USB package export
🎛️ Runtime config/config.ini control for vibe_level + god_mode + dry_run

Voice command list (offline parser)

status
stop
export
target add <X>
crack wifi
full auto
god mode
make legendary

Vibe levels

Set in config/config.ini:

[runtime]
vibe_level = stealth     ; stealth | aggressive | cinematic
god_mode = false
dry_run = true

God mode

god mode switches to simulation-first flow for demos:

fake rapid results
matrix/glitch OLED visuals
safety-preserving no-live-aggression defaults

Physical hardware checklist

Raspberry Pi Zero 2 W boots and reaches shell
Whisplay OLED active + matrix boot animation visible
Button A triggers listen cycle
Button B short press cycles Bud personality
Button B long press performs emergency stop
PiSugar battery values visible in HUD
python main.py export writes encrypted package to mounted USB

Built 2026 by madman + Cursor.

🏗️ Project Structure

Omni-Bot/
├── README.md                    # This file
├── README_JARVIS_BUD.md        # Comprehensive Jarvis-Bud guide
├── LICENSE                      # MIT License
├── requirements.txt             # Python dependencies
├── setup.sh                     # Automated RPi setup script
│
├── jarvis_bud/                 # Main application
│   ├── main.py                # Entry point with async loop
│   ├── config.py              # Configuration management
│   ├── buds.json              # Personality definitions
│   │
│   ├── hardware/
│   │   ├── lcd.py            # ST7789 display driver
│   │   ├── audio.py          # WM8960 codec driver
│   │   ├── battery.py        # PiSugar 3 battery manager
│   │   └── buttons.py        # GPIO button handler
│   │
│   ├── ui/
│   │   ├── renderer.py       # PIL frame renderer
│   │   ├── animations.py     # Waveform animations
│   │   └── themes.py         # UI color schemes
│   │
│   ├── ai/
│   │   ├── ollama_client.py      # Local LLM (Ollama)
│   │   ├── openrouter_client.py  # Cloud LLM (OpenRouter)
│   │   └── router.py             # Intelligent routing
│   │
│   └── wizard/
│       └── setup_wizard.py      # First-boot setup
│
└── config/
    └── example_config.json      # Config template

🎯 Core Architecture

Event Loop Design (60fps)

┌─ Async Main Loop ──────────────────────────────┐
├─ Display Update (60fps)                        │
├─ Button Event Handler                          │
├─ AI Response Generation (streaming)            │
├─ Battery Monitoring                            │
└────────────────────────────────────────────────┘

Dual-Route AI

User Input
    ↓
AIRouter decides:
    ├─ Local Ollama (fast, offline)      [home/office]
    └─ Cloud OpenRouter (powerful)       [mobile/outdoor]
    ↓
AI Response → Waveform Animation → Output

Hardware Stack

Component	Driver	Purpose
RPi Zero 2W	Custom	Main CPU
ST7789 LCD	PIL-based	User Interface
WM8960 Codec	alsaaudio/pyaudio	Audio I/O
PiSugar 3	Socket IPC	Battery + UPS
GPIO Buttons	RPi.GPIO	Physical Input

🚀 Getting Started

Prerequisites

Raspberry Pi Zero 2 W (or any RPi with SPI/I2C support)
Whisplay HAT with WM8960 codec
ST7789 LCD display (240x280)
PiSugar 3 battery management
Python 3.8+
WiFi/Ethernet for cloud AI (optional)

Installation (5 min)

Clone this repo:

git clone https://github.com/NaustudentX18/Omni-Bot.git
cd Omni-Bot

Run automated setup:
```
bash setup.sh
```
This handles: system updates, Python dependencies, SPI/I2C enabling, config directory creation
Start Jarvis-Bud:
```
python3 -m jarvis_bud.main
```
Follow the interactive wizard:
- Step 1: Hardware Check ✅
- Step 2: WiFi & Ollama auto-detection 📡
- Step 3: Pick your personality 🤝
- Step 4: Add OpenRouter API (optional) 🔑
Done! Your AI companion is ready 🎉

💡 Usage

Physical Controls

Button A (GPIO 17): Press to activate Listen mode 🎙️
Button B (GPIO 27): Press to cycle through Buds 💫

Voice Interaction

1. Press Button A
2. Speak your question
3. Release Button A
4. Watch waveform animate while thinking
5. Hear AI response through speaker

LCD Display

🔋 85%                    🏠  ← Battery & Connectivity Status
─────────────────────────────
    📍 Tech-Support Bud 🤖
    
    ▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄  ← Waveform Animation
    
─────────────────────────────
    Local Ollama

🔧 Configuration

First-Time Setup

The wizard creates config/config.json with your preferred settings

Manual Configuration

Edit config/config.json:

{
  "bud": {
    "id": "tech-support",
    "name": "Tech-Support Bud 🤖",
    "system_prompt": "You are a friendly tech-support assistant...",
    "ui_accent_color": "#00FF00"
  },
  "connectivity": {
    "ollama": "http://naspi:11434"
  },
  "openrouter": {
    "api_key": "sk_your_key_here",
    "model": "google/gemini-2.0-flash-lite:free"
  }
}

Adding AI Backends

Local Ollama (Free, Offline):

# On your home server
ollama pull mistral
ollama serve
# Pi will auto-detect at common addresses

Cloud OpenRouter (Powerful, Mobile):

1. Get free API key: https://openrouter.ai/keys
2. Add to wizard at setup or config.json
3. Enjoy cloud AI models (Gemini, Llama, etc.)

🎨 UI & Personality

Built-In Personalities

Tech-Support Bud 🤖 - Green neon, patient & knowledgeable
Trail-Runner Bud 🏃 - Blue neon, adventurous & motivational
Snarky Hacker Bud 🖥️ - Magenta neon, witty & irreverent
Zen-Guide Bud 🧘 - Orange neon, calm & philosophical

Customize Appearance

Edit jarvis_bud/buds.json to add new personalities with custom:

System prompts (behavior)
UI accent colors
Response styles
Personality traits

Theme Colors

Terminal-Dark with Neon Accents:

Background: #0A0A0F (deep blue-black)
Text: #D2DCE6 (light gray)
Neon Green: #00FF64 (primary accent)
Neon Blue: #0096FF (secondary accent)

🧠 AI Models

Recommended Models

Local (Ollama) - Zero cost, instant, offline

mistral — 4.1B params, lightweight
neural-chat — Optimized for conversation
starling-lm — 7B, highly capable

Cloud (OpenRouter) - Powerful, anywhere, free tier

google/gemini-2.0-flash-lite:free — Fast & smart
meta-llama/llama-3.3-70b-instruct:free — Most powerful
mistralai/mixtral-8x7b-instruct:free — Balanced

📚 API Reference

Hardware

from jarvis_bud.hardware import ST7789Display, AudioCodec, Battery, ButtonHandler

lcd = ST7789Display()
audio = AudioCodec()
battery = Battery()
buttons = ButtonHandler()

UI

from jarvis_bud.ui import FrameRenderer, WaveformAnimator

renderer = FrameRenderer()
animator = WaveformAnimator()
animator.get_frame("idle")  # idle, listening, thinking, loading

AI

from jarvis_bud.ai import AIRouter

router = AIRouter(
    ollama_url="http://localhost:11434",
    openrouter_key="sk_..."
)
response = router.generate("Question?", prefer="auto")

Configuration

from jarvis_bud.config import ConfigManager

config = ConfigManager()
bud_name = config.get("bud.name")
prompt = config.get_bud_system_prompt()
config.set("bud.id", "snarky-hacker")
config.save()

📖 Full API Documentation →

🐛 Troubleshooting

Common Issues

LCD not displaying?

Check SPI enabled: sudo raspi-config → SPI → Enable
Verify backlight GPIO (usually GPIO 12)

Audio not working?

Check codec: i2cdetect -y 1 (look for WM8960)
Unmute: alsamixer (space to unmute)

Ollama not found?

Start Ollama server on your home PC/server
Verify same WiFi network
Check custom IP in config.json

Stuck in setup wizard?

Hold Button A for 10 seconds to force-boot
Delete config.json to re-run wizard: rm config/config.json

📖 Full Troubleshooting Guide →

🤝 Contributing

We'd love your help! Possible contributions:

🎭 New Bud personalities with unique system prompts
🎨 UI themes and visual customizations
🔌 Additional hardware driver support
🎬 Animation effects and transitions
📖 Documentation improvements
🐛 Bug reports and testing

Development

# Clone and setup
git clone https://github.com/NaustudentX18/Omni-Bot.git
cd Omni-Bot
python3 -m venv venv && source venv/bin/activate
pip install -r requirements.txt

# Code quality
black jarvis_bud/
pylint jarvis_bud/

# Test
python3 -m pytest tests/

📜 License

MIT License - See LICENSE file for details

🎓 Acknowledgments

Jarvis-Bud brings together:

Raspberry Pi ecosystem & community
Ollama local LLM technology
OpenRouter for cloud AI
PIL/Pillow for UI rendering
asyncio for responsive design

📞 Links & Resources

Jarvis-Bud Full Guide: README_JARVIS_BUD.md
GitHub: https://github.com/NaustudentX18/Omni-Bot
Ollama: https://ollama.ai (Local LLM)
OpenRouter: https://openrouter.ai (Cloud AI)
Raspberry Pi Docs: https://www.raspberrypi.com/documentation/

🎯 Roadmap

Roadmap completion status: ✅ Complete (10/10 implemented)

🚀 Your AI companion awaits. Let's build the future together!

     ^
    / \__
   (    @\___
   /         O
  /   (_____/
/_____/   U

Jarvis-Bud: Edge AI for makers, hikers, and tech-heads

Name		Name	Last commit message	Last commit date
config		config
jarvis_bud		jarvis_bud
nxtgenai		nxtgenai
scripts		scripts
tests		tests
.gitignore		.gitignore
AGENTS.md		AGENTS.md
LICENSE		LICENSE
README.md		README.md
README_JARVIS_BUD.md		README_JARVIS_BUD.md
ai_core.py		ai_core.py
main.py		main.py
pyproject.toml		pyproject.toml
requirements-dev.txt		requirements-dev.txt
requirements.txt		requirements.txt
setup.sh		setup.sh
tools.py		tools.py
ui.py		ui.py

Latest commit

History

Folders and files

Repository files navigation