Run a multi-platform AI assistant on your own machine — with full privacy.

Telegram · Discord · WhatsApp · Slack — powered by open-source LLMs, 100% offline.

Se você acha privacidade importante, este projeto é para você. If you care about privacy, this project is for you.

• 🔒 Zero cloud dependency — your conversations never leave your machine
• 🧠 Real local AI — not a wrapper, a full local LLM runtime via Ollama
• 📱 Works where you already are — Telegram, WhatsApp, Discord, Slack
• 🌐 Bilingual — English + Portuguese, more languages welcome
• 🛠️ Open & extensible — MIT license, clean architecture, easy to hack
• 🚀 Active roadmap — Web UI, Docker one-click, Plugin system coming

⭐ Star the repo to support open-source, private AI — it takes 2 seconds and means the world to the maintainer.

• 🇺🇸 English
• 🇧🇷 Português

Boot	Telegram	AI Chat
CrustAI booting up	Bot live on Telegram	AI responding locally

CrustAI is a self-hosted AI assistant that runs 100% locally using Ollama. It integrates with Telegram, Discord, WhatsApp and Slack so you can chat with your AI in tools you already use — without sending your data to any cloud LLM provider.

Built with Node.js and powered by Ollama (local LLM runtime), CrustAI is designed for developers, privacy enthusiasts, and anyone who wants an AI assistant that truly belongs to them.

• Node.js ≥ 20.0
• Ollama installed and running
• A Telegram Bot Token from @BotFather

# 1. Clone the repository
git clone https://github.com/DaveSimoes/CrustAI.git
cd CrustAI

# 2. Install dependencies
npm install

# 3. Start Ollama and pull a model
ollama serve
ollama pull tinyllama   # lightweight (600MB)
# or
ollama pull llama3.2    # more powerful (2GB, needs 8GB RAM)

# 4. Configure the project
cp config/config.example.yml config/config.yml
# Edit config/config.yml with your Telegram token and model

# 5. Run CrustAI
npm start

Edit config/config.yml:

model: tinyllama          # or llama3.2, phi3, mistral...
ollama_url: http://localhost:11434
language: pt-BR

telegram:
  enabled: true
  token: YOUR_BOT_TOKEN_HERE
  allowed_user_ids: []    # leave empty to allow all users

discord:
  enabled: false
  token: ""

whatsapp:
  enabled: false

voice:
  enabled: false
  port: 8765

/ping      → Check if the bot is alive
/help      → Show all commands
/model     → Show which AI model is running
/remember  → Store a fact in long-term memory
/forget    → Erase all stored facts
/clear     → Clear conversation history

Adapters (Telegram / Discord / WhatsApp / Slack)
                │
                ▼
         Message Orchestrator
        ┌────────┴────────┐
        ▼                 ▼
   Ollama Client      Memory Store
        │                 │
        └────────┬────────┘
                 ▼
             REST API

Design note: adapter boundaries make it easy to add new channels without changing core conversation logic.

crustai/
├── src/
│   ├── core/
│   │   ├── index.js        # Main orchestrator
│   │   ├── llm.js          # Ollama LLM client
│   │   └── commands.js     # Command handler
│   ├── adapters/
│   │   ├── telegram/       # Telegram bot
│   │   ├── discord/        # Discord bot
│   │   ├── whatsapp/       # WhatsApp bot
│   │   └── slack/          # Slack bot
│   ├── memory/
│   │   └── store.js        # Long-term memory
│   ├── personality/
│   │   └── prompt.js       # System prompt builder
│   ├── voice/
│   │   └── server.js       # Voice WebSocket server
│   └── api/
│       └── server.js       # REST API
├── config/
│   ├── config.yml          # Your configuration (git-ignored)
│   ├── config.example.yml  # Template
│   └── personality.yml     # Assistant personality
├── demo/
│   ├── terminal.gif        # Boot demo
│   ├── ping.gif            # Telegram connection demo
│   └── chat.gif            # AI conversation demo
└── data/                   # Local database (git-ignored)

CrustAI was built with privacy as its core principle:

• ✅ All conversations stay on your machine
• ✅ No API keys sent to external AI services
• ✅ No telemetry or usage tracking
• ✅ Open source — inspect every line of code
• ✅ Your data, your rules

• 🖥️ Web UI dashboard
• 🖼️ Image understanding (multimodal LLMs)
• 🧩 Plugin system for custom tools
• 🐳 Docker one-click deployment
• 📱 Mobile app companion

💡 Want to help build any of these? Check the open issues — contributions are very welcome!

Contributions, issues and feature requests are welcome!

1. Fork the project
2. Create your feature branch: git checkout -b feat/my-feature
3. Commit your changes: git commit -m 'feat: add my feature'
4. Push to the branch: git push origin feat/my-feature
5. Open a Pull Request

See CONTRIBUTING.md for detailed guidelines.

Dave Simoes — Developer passionate about AI, privacy and open source.

• 🐙 GitHub: @DaveSimoes
• 💼 LinkedIn: Dave Simoes

⭐ Support the Project

If CrustAI helped you, please give it a star — it helps a lot! Se o CrustAI te ajudou, deixe uma estrela — faz toda a diferença!

Made with 🦀 and ❤️ by Dave Simoes

CrustAI é um assistente de IA totalmente privado e auto-hospedado que roda inteiramente na sua própria máquina — nenhum dado sai do seu computador. Ele se conecta a plataformas de mensagens populares como Telegram, WhatsApp, Discord e Slack, oferecendo o poder de uma IA conversacional sem abrir mão da sua privacidade.

Construído com Node.js e alimentado pelo Ollama (motor de LLM local), o CrustAI foi projetado para desenvolvedores, entusiastas de privacidade e qualquer pessoa que queira um assistente de IA que realmente lhe pertença.

# 1. Clone o repositório
git clone https://github.com/DaveSimoes/CrustAI.git
cd CrustAI

# 2. Instale as dependências
npm install

# 3. Inicie o Ollama e baixe um modelo
ollama serve
ollama pull tinyllama

# 4. Configure o projeto
cp config/config.example.yml config/config.yml
# Edite config/config.yml com seu token do Telegram

# 5. Inicie o CrustAI
npm start

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

Este projeto está licenciado sob a Licença MIT — veja o arquivo LICENSE para detalhes.