n8n AI-Powered Telegram Bot with OpenRouter

This n8n workflow creates a witty and engaging Telegram bot that leverages AI to generate humorous roasts and provide user statistics. It uses the OpenRouter platform for AI language models, allowing for flexible and powerful conversational capabilities.

The bot listens for messages in Telegram, processes them through an AI agent, and can respond with a roast or provide user chat statistics, all while storing interaction data in a PostgreSQL database.

What it does

1. Listens for Telegram Messages: The workflow is triggered by incoming messages to a configured Telegram bot.
2. Filters Bot Commands: It checks if the incoming message is a specific command (e.g., /roast or /stats).
3. Handles /roast Command:
   • If the command is /roast, it extracts the target of the roast from the message.
   • It then uses an AI Agent (LangChain with OpenRouter) to generate a humorous roast based on the provided target.
   • The generated roast is sent back to the user in Telegram.
4. Handles /stats Command:
   • If the command is /stats, it queries a PostgreSQL database to retrieve interaction statistics for the user.
   • The statistics are formatted and sent back to the user in Telegram.
5. Stores User Interactions: All incoming messages are logged into a PostgreSQL database, allowing for tracking and analysis of bot usage.
6. Scheduled Database Cleanup (Optional but Recommended): A separate branch of the workflow is triggered on a schedule (e.g., daily) to perform database cleanup or maintenance tasks. (Note: The provided JSON only shows the trigger, the actual cleanup logic would need to be added).

Prerequisites/Requirements

• n8n Instance: A running n8n instance (self-hosted or cloud).
• Telegram Bot: A Telegram bot token obtained from BotFather.
• OpenRouter Account & API Key: An OpenRouter API key for accessing AI language models.
• PostgreSQL Database: Access to a PostgreSQL database for storing user interaction data.
• n8n LangChain Nodes: Ensure the @n8n/n8n-nodes-langchain package is installed in your n8n instance.

Setup/Usage

1. Import the Workflow:
   • Download the provided JSON file.
   • In your n8n instance, go to "Workflows" and click "New".
   • Click the "Import from JSON" button and paste the workflow JSON or upload the file.
2. Configure Credentials:
   • Telegram Trigger & Telegram Node: Create or select an existing Telegram API credential using your bot token.
   • OpenRouter Chat Model: Create or select an OpenRouter API credential using your OpenRouter API Key.
   • Postgres Node: Create or select a PostgreSQL credential with the necessary connection details (host, port, database, user, password).
3. Database Schema:
   • Ensure your PostgreSQL database has a table to store Telegram messages. A simple schema might look like:

     CREATE TABLE telegram_messages (
         id SERIAL PRIMARY KEY,
         chat_id VARCHAR(255),
         user_id VARCHAR(255),
         username VARCHAR(255),
         message_text TEXT,
         timestamp TIMESTAMP DEFAULT CURRENT_TIMESTAMP
     );
     

   • Adjust the Postgres node's query to match your table and column names if different.
4. Activate the Workflow:
   • After configuring all nodes and credentials, activate the workflow by toggling the "Active" switch in the top right corner of the workflow editor.
5. Interact with the Bot:
   • Open Telegram and send messages to your bot.
   • Try commands like /roast <name> (e.g., /roast n8n) or /stats to see it in action.
6. Customize AI Agent:
   • Adjust the AI Agent node's prompt and tools to refine the bot's personality and capabilities.
   • Experiment with different OpenRouter models to find the best balance of humor and performance.
7. Scheduled Cleanup (Optional):
   • Configure the Schedule Trigger node (ID 839) to your desired interval for database maintenance.
   • Add the necessary Postgres nodes and logic after the Schedule Trigger to perform tasks like deleting old messages or generating reports.