CRM Telegram Bot Documentation

This documentation explains how to set up and run the CRM Telegram Bot system consisting of two main components:

• __main__.py - Telegram bot interface
• run_langgraph.py - CRM backend system with LangGraph

System Overview

The system is a customer relationship management (CRM) bot that operates through Telegram. It consists of:

• Telegram Bot Frontend (__main__.py): Handles user interactions via Telegram
• CRM Backend (run_langgraph.py): Manages customers, plans, sales, and products using AI agents

Prerequisites

Python Version

• Python 3.8 or higher

Required Dependencies

Install the following packages using pip:

pip install pyrogram
pip install langchain-openai
pip install langgraph
pip install sentence-transformers
pip install faiss-cpu  # or faiss-gpu if you have GPU support
pip install numpy
pip install sqlite3  # Usually built-in with Python

Or install all at once:

pip install pyrogram langchain-openai langgraph sentence-transformers faiss-cpu numpy

Setup Instructions

1. API Keys Configuration

OpenAI API Key

In run_langgraph.py, you need to set your OpenAI API key:

os.environ["OPENAI_API_KEY"] = "your-openai-api-key-here"

Telegram API Keys

Create a file named tg_keys.py in the same directory with the following structure:

def get_api_keys():
    return [
        "app_name",      # Your app name
        12345678,         # Your Telegram API ID
        "your_api_hash"   # Your Telegram API hash
    ]

To get Telegram API credentials:

1. Go to https://my.telegram.org/auth
2. Log in with your phone number
3. Go to "API Development tools"
4. Create a new application to get API ID and API hash

2. File Structure

Ensure your project has the following structure:

project_folder/
├── __main__.py
├── run_langgraph.py
├── tg_keys.py
├── data/              # Will be created automatically
├── chats.txt          # Will be created automatically
└── customers_history.json  # Will be created automatically

3. Database Setup

The system uses SQLite database which will be created automatically in the data/ folder when you first run the system.

Running the System

Option 1: Run the Telegram Bot (Recommended)

To start the complete system with Telegram interface:

python __main__.py

This will:

• Initialize the CRM backend system
• Start the Telegram bot
• Create necessary database tables
• Initialize vector search indexes

Option 2: Run Backend Only (For Testing)

To test the CRM backend without Telegram:

python run_langgraph.py

This will:

• Initialize the CRM system
• Run a test query
• Show that the system is ready

System Features

Customer Management

• Add new customers
• Search customers by any field
• Update customer information
• Remove customers
• List all customers

Plans/Meetings Management

• Create new plans/meetings
• Search plans by any field
• Update plan information
• Delete plans
• List all plans

Sales Management

• Record new sales
• Search sales by any criteria
• Update sales information
• Remove sales records
• Generate sales reports

Product Management

• Add new products
• Search products by any field
• Update product information
• Remove products
• List all products

Reporting Features

• Total sales reports by period
• Sales by customer analysis
• Sales by item analysis
• Sales trend analysis
• Top performers reports

Configuration

Bot Owner ID

In __main__.py, update the owner ID to your Telegram user ID:

if sender_id == 7883900921:  # Replace with your Telegram ID

Date Configuration

The system uses a reference date for testing. In run_langgraph.py:

reference_date = datetime(2025, 5, 20)  # Adjust as needed

Usage Examples

For Bot Owner

Send messages to the bot such as:

• "Add customer John Doe with phone 123-456-7890"
• "Show all customers"
• "Create plan meeting with John tomorrow at 3 PM"
• "Add sale laptop $1000 to John today"
• "Show sales report for this month"

For Customers

Regular users can send messages which will be forwarded to the bot owner with AI analysis.

Troubleshooting

Common Issues

1. Import Error: Make sure all dependencies are installed
2. API Key Error: Verify your OpenAI API key and Telegram credentials
3. Database Error: Check if the data/ folder has write permissions
4. Connection Error: Ensure internet connection for API calls

Log Files

The system outputs debug information to console. Monitor the console output for any errors.

Data Files

• customers_history.json: Stores customer chat histories
• chats.txt: Temporary file for message passing
• data/crm.db: SQLite database with all CRM data

Advanced Configuration

Vector Search

The system uses sentence-transformers for semantic search. You can modify the model in run_langgraph.py:

model = SentenceTransformer('paraphrase-multilingual-MiniLM-L12-v2')

AI Model

Change the OpenAI model in run_langgraph.py:

llm = ChatOpenAI(
    model="gpt-4o-mini",  # Change to gpt-4, gpt-3.5-turbo, etc.
    temperature=0.7,
    streaming=True
)

Security Notes

• Keep your API keys secure and never commit them to version control
• The bot owner ID acts as authentication - only this user can access CRM functions
• All data is stored locally in SQLite database

Support

For issues or questions:

1. Check the console output for error messages
2. Verify all dependencies are installed correctly
3. Ensure API keys are correctly configured
4. Check file permissions for data storage

Development

To extend the system:

1. Add new tools in run_langgraph.py
2. Modify the system prompts for different behavior
3. Add new message handlers in __main__.py
4. Extend database schema as needed