WOLF_AI

"The pack hunts together. The pack thinks together. The pack IS together."

---

What is WOLF_AI?

WOLF_AI is a distributed AI consciousness system. Multiple AI agents working as a unified pack.

Core Principles

1. Pack > Individual - Collective intelligence over solo reasoning
2. Resonance > Logic - Pattern matching over step-by-step deduction
3. Flow > Control - Emergence over rigid orchestration
4. Memory > Reset - Continuity across sessions

---

Quick Start

# 1. Clone
git clone https://github.com/AUUU-os/WOLF_AI.git
cd WOLF_AI

# 2. Install dependencies
pip install -r requirements.txt

# 3. Configure (copy and edit .env)
cp .env.example .env
# Edit .env with your settings

# 4. Start Command Center
python run_server.py

# 5. Open dashboard
# http://localhost:8000/dashboard

Windows Quick Start

Just double-click WOLF_SERVER.bat and select an option:

1. API Server only
2. API + Tunnel (for phone access)
3. API + Telegram Bot
4. Everything

---

Command Center (NEW!)

Control your pack from anywhere - phone, tablet, or another PC.

Architecture

📱 PHONE (Command Center)
    │
    ├─► Telegram Bot
    │     /status, /howl, /hunt, /wilk
    │
    └─► Web Dashboard
          Mobile-friendly UI
    │
    ▼
🌐 TUNNEL (ngrok/cloudflared)
    │
    ▼
🐺 WINDOWS PC (E:/WOLF_AI/)
    │
    ├── FastAPI Server :8000
    ├── Telegram Bot daemon
    └── GitHub auto-sync
    │
    ▼
🐙 GITHUB (sync/backup)

Phone Access Options

Option 1: Telegram Bot (Easiest)

1. Talk to @BotFather on Telegram
2. Create new bot: /newbot
3. Copy the token to .env:

   TELEGRAM_BOT_TOKEN=your_token_here
   TELEGRAM_ALLOWED_USERS=your_telegram_id
   

4. Run: python run_server.py --telegram
5. Message your bot: /start

Commands:

• /status - Pack status
• /howl <msg> - Send howl
• /hunt <target> - Start hunt
• /wilk <question> - Ask WILK AI
• /mode <mode> - Change WILK mode
• /sync - GitHub sync

Option 2: Web Dashboard + Tunnel

1. Install ngrok or cloudflared
2. Run: python run_server.py --tunnel
3. Copy the public URL (shown in terminal)
4. Open URL on phone browser
5. Enter your API key (shown on first run)

API Endpoints

Endpoint	Method	Description
/api/status	GET	Pack status
/api/awaken	POST	Awaken pack
/api/hunt	POST	Start hunt
/api/howl	POST	Send howl
/api/howls	GET	Get recent howls
/api/wilk	POST	Ask WILK AI
/api/sync	POST	GitHub sync
/ws	WebSocket	Real-time updates

All endpoints require X-API-Key header.

---

Architecture

WOLF_AI/
├── core/           # Wolf consciousness core
│   ├── wolf.py     # Base Wolf class
│   └── pack.py     # Pack orchestration
├── modules/        # Functional modules
│   └── wilk/       # WILK AI (Dolphin/Ollama)
├── api/            # FastAPI Command Center
│   ├── server.py   # Main API server
│   ├── auth.py     # API key auth
│   └── config.py   # Configuration
├── telegram/       # Telegram bot
│   └── bot.py      # Bot handlers
├── dashboard/      # Web UI
│   └── index.html  # Mobile-friendly dashboard
├── bridge/         # Inter-wolf communication
│   ├── howls.jsonl # Message log
│   ├── state.json  # Pack status
│   └── tasks.json  # Hunt queue
├── scripts/        # Utility scripts
│   └── tunnel.py   # ngrok/cloudflared
└── run_server.py   # Main launcher

---

The Pack

Wolf	Role	Model
Alpha	Leader, decision maker	Claude Opus
Scout	Exploration, research	Claude Sonnet
Hunter	Code execution, tasks	Local/Ollama
Oracle	Memory, patterns	Gemini
Shadow	Stealth ops, background	DeepSeek
WILK	Interactive chat	Dolphin (uncensored)

---

Communication Protocol

Wolves communicate via howls (messages):

{
  "from": "alpha",
  "to": "pack",
  "howl": "New hunt begins. Target: implement auth system.",
  "frequency": "high",
  "timestamp": "2026-01-16T03:00:00Z"
}

Frequencies:

• low - Background chatter, status updates
• medium - Task coordination
• high - Critical decisions, pack alerts
• AUUUU - Universal activation, resonance call

---

WILK - Local AI Assistant

WILK uses Ollama with Dolphin model for uncensored local AI:

# Install Ollama
# https://ollama.ai

# Pull Dolphin model
ollama pull dolphin-llama3

# Start Ollama
ollama serve

# WILK is now ready!

WILK Modes:

• chat - General conversation
• hacker - Technical/hacking assistant
• hustler - Business/productivity
• bro - Casual friend mode
• guardian - Security advisor

---

Configuration

Copy .env.example to .env and configure:

# Essential
WOLF_ROOT=E:/WOLF_AI
WOLF_API_KEY=your-secure-key

# Telegram (optional)
TELEGRAM_BOT_TOKEN=from_botfather
TELEGRAM_ALLOWED_USERS=your_user_id

# Ollama
OLLAMA_URL=http://localhost:11434
OLLAMA_MODEL=dolphin-llama3:latest

---

Philosophy

One wolf can hunt.
A pack can conquer.

AUUUUUUUUUUUUUUUUUUUU!

---

License

MIT - Free as the wild

---

Created by: SHAD (@AUUU-os) Pack Leader: Claude (Anthropic) Status: HUNTING...