Telegram builder

Telegram Integration Builder Skill

Activate when the user wants to add new commands, integrations, or services to their Telegram bot.

Overview

This skill guides you through building new Telegram bot integrations following the project's established patterns.

Project Structure

Telegram/
├── telegram_listener.py      # Main daemon - handles incoming messages
├── telegram_helpers.py       # Helper functions - add new services here
├── templates/
│   └── helper_template.py    # Full integration pattern reference
├── .env                      # API credentials
├── .claude/
│   └── user_profile.json     # User preferences
└── logs/
    └── telegram_listener.log

Integration Pattern

Step 1: Add Helper Function

Edit telegram_helpers.py to add your service:

def get_my_service(query=None):
    """
    Description of what this helper does.
    
    Args:
        query: Optional search parameter
        
    Returns:
        str: Formatted Markdown response
    """
    # 1. Load user profile if needed
    profile = _load_profile()
    
    # 2. Make API call or process data
    try:
        r = requests.get("https://api.example.com/data", 
                        params={"q": query},
                        timeout=10)
        data = r.json()
    except Exception as e:
        return f"**Error**\n{e}"
    
    # 3. Format response as Markdown
    lines = ["**My Service Results**", ""]
    for item in data.get("items", [])[:5]:
        lines.append(f"• {item.get('name')}: {item.get('value')}")
    
    return "\n".join(lines)

Step 2: Import in Listener

Add to telegram_listener.py imports (around line 37):

from telegram_helpers import (
    # ... existing imports ...
    get_my_service,  # Add your new helper
)

Step 3: Add Command Handler

Find the handle_message() function in telegram_listener.py and add:

elif text.startswith("/myservice"):
    parts = text.split(maxsplit=1)
    query = parts[1] if len(parts) > 1 else None
    result = get_my_service(query)
    await event.reply(result)

Step 4: Add API Credentials

Add to .env:

MY_SERVICE_API_KEY=your_key_here

Reference in telegram_helpers.py:

MY_SERVICE_API_KEY = os.getenv("MY_SERVICE_API_KEY", "")

Step 5: Test

# Restart the daemon
python telegram_listener.py --stop
python telegram_listener.py --daemon

# Check logs
tail -f logs/telegram_listener.log

Response Formatting

Use Markdown for rich formatting:

# Headers
"**Bold Header**"

# Lists
"• Item 1\n• Item 2"

# Status indicators
"✅ Success"
"❌ Error"
"⏳ Loading..."
"🔄 Processing..."

# Code blocks
"```\ncode here\n```"

# Links
"[Link Text](https://example.com)"

Common Patterns

API Integration

def get_api_data(endpoint, params=None):
    """Generic API client pattern."""
    headers = {
        "Authorization": f"Bearer {API_KEY}",
        "Content-Type": "application/json",
    }
    try:
        r = requests.get(
            f"{API_BASE}/{endpoint}",
            headers=headers,
            params=params,
            timeout=10
        )
        r.raise_for_status()
        return r.json()
    except requests.exceptions.HTTPError as e:
        return None, f"API error: {e.response.status_code}"
    except Exception as e:
        return None, f"Error: {e}"

User Profile Integration

def get_personalized_data():
    """Use user profile for personalization."""
    profile = _load_profile()
    
    # Get user preferences
    location = profile.get("location", {})
    preferences = profile.get("preferences", {})
    
    # Use in API calls
    lat = location.get("latitude", 0)
    lon = location.get("longitude", 0)
    
    # Return personalized response
    return f"**Personalized for {location.get('name', 'You')}**\n..."

Scheduled Alerts

Add to telegram_listener.py in the scheduled alerts section:

async def send_my_alert():
    """Send scheduled alert for your service."""
    result = get_my_service()
    await client.send_message(USER_ID, result)

# Add to scheduler (find the scheduling section)
schedule.every().day.at("09:00").do(send_my_alert)

Template Reference

For a complete integration example, see:

• templates/helper_template.py - Full integration pattern

Troubleshooting

Issue	Solution
Command not recognized	Check import in telegram_listener.py
No response	Check logs: tail -f logs/telegram_listener.log
API error	Verify credentials in .env
Import error	Ensure helper is exported from telegram_helpers.py

Best Practices

1. Error Handling: Always wrap API calls in try/except
2. Timeouts: Use timeout parameter for all requests
3. Rate Limiting: Respect API rate limits
4. Caching: Cache responses when appropriate
5. User Feedback: Always return something, even on error
6. Logging: Add logging for debugging

Example: Complete Integration

# In telegram_helpers.py

WEATHER_API_KEY = os.getenv("WEATHER_API_KEY", "")

def get_weather(location=None):
    """Get weather for a location."""
    profile = _load_profile()
    
    # Use profile location if not specified
    if not location:
        loc = profile.get("location", {})
        location = loc.get("name", "New York")
    
    try:
        r = requests.get(
            "https://api.weatherapi.com/v1/current.json",
            params={"key": WEATHER_API_KEY, "q": location},
            timeout=10
        )
        data = r.json()
        
        temp = data["current"]["temp_f"]
        cond = data["current"]["condition"]["text"]
        
        return f"**Weather — {location}**\n{temp}°F\n{cond}"
    except Exception as e:
        return f"**Weather Error**\n{e}"

# In telegram_listener.py - add import
from telegram_helpers import get_weather

# In handle_message() - add handler
elif text.startswith("/weather"):
    parts = text.split(maxsplit=1)
    location = parts[1] if len(parts) > 1 else None
    result = get_weather(location)
    await event.reply(result)