Quick Start
Get your Teleton Agent running in 5 minutes with this step-by-step guide.
Prerequisites
Before starting, make sure you have:
- Node.js 20+ installed (
node --version) - A Telegram account with a phone number
- An API key from Anthropic, OpenAI, or another supported provider
- Telegram API credentials from my.telegram.org
Step 1: Install Teleton
Terminal
npm install -g teleton@latest
Step 2: Run the Setup Wizard
The setup wizard will guide you through configuration:
Terminal
teleton setup
The wizard will ask for:
- Telegram API ID and Hash - Get from my.telegram.org
- Phone number - Your Telegram account's phone
- Verification code - Sent to your Telegram
- LLM provider - Choose Anthropic, OpenAI, etc.
- API key - Your provider's API key
Telegram API credentials
Visit my.telegram.org, log in, go to "API development tools", and create an application to get your API ID and Hash.
Step 3: Start the Agent
Terminal
teleton start
# Or with WebUI dashboard:
teleton start --webui
You should see:
Output
✓ Connected to Telegram as @YourUsername
✓ Loaded 114 tools
✓ Memory system initialized
✓ Agent is ready and listening...
Step 4: Test Your Agent
Send a message to your Telegram account from another device or user. Try:
Telegram Message
Hello! What can you do?
Your agent should respond with its capabilities. You can also test admin commands:
| Command | What it does |
|---|---|
/status |
Show agent status and memory usage |
/wallet |
Display your TON wallet address |
/model |
Show current LLM model |
/pause |
Pause the agent |
/resume |
Resume the agent |
Optional: Enable WebUI
Access the dashboard at http://localhost:3000 when running with --webui:
Terminal
teleton start --webui
# Dashboard available at http://localhost:3000
The WebUI provides:
- Real-time message monitoring
- Memory and conversation history
- Tool usage statistics
- Configuration management
Common Issues
Authentication Failed
If Telegram authentication fails:
- Ensure your phone number includes country code (
+1,+44, etc.) - Check that your API ID/Hash are correct
- Delete
~/.teleton/session.jsonand try again
LLM API Errors
If you see API errors:
- Verify your API key is valid and has credits
- Check your network connection
- Try a different model with
/modelcommand
Memory/Database Issues
If memory system fails to initialize:
- Ensure
~/.teletondirectory is writable - Check disk space
- Try
teleton reset --memoryto rebuild the database