Workflow Link: https://aniketgund12.app.n8n.cloud/workflow/EvUmoUezPd6Aog3M
Telegram Link: https://t.me/Griffolyon_bot
This project is an AI-powered Telegram chatbot built using n8n workflow automation. The bot listens to user messages on Telegram, processes them through an AI Agent powered by Google Gemini, optionally retains conversation context using memory, and sends intelligent responses back to the user in real time.
The main goal of this project is to demonstrate how low-code automation platforms like n8n can be used to build production-ready AI agents without writing a full backend from scratch.
The chatbot workflow consists of the following components:
- Telegram Trigger
Listens for incoming messages from a Telegram bot
Acts as the entry point of the workflow
Trigger type: message
- AI Agent (Core Logic)
Receives user input from Telegram
Uses an LLM to generate intelligent responses
Handles:
Prompt understanding
Context-aware replies
Conversational flow
- Chat Model – Google Gemini
Integrated using n8n’s AI node
Responsible for:
Natural language understanding
Response generation
Chosen for fast inference and strong conversational capabilities
- Memory Module (Optional but Implemented)
Stores past interactions
Enables:
Context-aware conversations
Follow-up questions
Demonstrates real-world chatbot behavior instead of stateless replies
- Telegram Send Message
Sends the AI-generated response back to the user
Completes the request–response cycle
User sends a message on Telegram
Telegram Trigger activates the workflow
Message is passed to the AI Agent
AI Agent processes input using Gemini model
Memory module retrieves or updates conversation history
Final response is generated
Response is sent back to the user via Telegram
n8n – Workflow automation platform
Telegram Bot API – User interaction interface
Google Gemini – AI chat model
AI Agent Node – Conversational intelligence
Memory Module – Context retention
Real-time Telegram chatbot
AI-generated intelligent responses
Context-aware conversations using memory
No-code / low-code implementation
Easily extendable (APIs, databases, tools can be added)
workflow.json – Exported n8n workflow
README.md – Project documentation
screenshots/ – Workflow and chatbot screenshots and video
Install or access n8n (cloud or self-hosted)
Import the provided workflow JSON
Create a Telegram bot using BotFather
Add Telegram credentials in n8n
Configure Google Gemini API credentials
Activate the workflow
Start chatting with your Telegram bot
AI customer support bots
Educational assistants
Personal productivity bots
FAQ automation
AI experimentation with minimal backend setup
Most beginners just build static rule-based bots. This project goes further by:
Using LLM-powered AI
Maintaining conversation memory
Applying real-world automation concepts
It shows clear understanding of:
AI workflows
Event-driven systems
Practical chatbot architecture
User authentication and role-based responses
Database-backed long-term memory
Tool calling (APIs, web search, internal tools)
Multi-language support
Deployment monitoring and logging