Telegram AI Chatbot with Long-Term Memory and Note Storage

Overview

This n8n workflow enables you to create an advanced AI chatbot for Telegram that goes beyond simple conversational responses. It utilizes Langchain to process user messages and integrate with LLMs, offering a sophisticated approach to AI interaction.

The core functionality revolves around two key memory management strategies:

1. Long-Term Memory: User-provided information that is deemed noteworthy and personal (preferences, habits, goals) is stored in Google Docs for future reference. This allows the AI to recall and use this information in subsequent conversations, personalizing the user experience.
2. Note Storage: Specific instructions, reminders, or standalone pieces of information that the user wishes to store are also managed, potentially via Google Docs or another designated method. This separates actionable items from general personal memories.

The workflow starts with a When chat message received trigger on Telegram, processes the input through an AI Tools Agent (which uses an LLM like GPT-4o-mini or DeepSeek-V3), and then decides whether to save information as memory or as a note. It also incorporates a Window Buffer Memory for short-term context and can optionally send responses back to Telegram.

This workflow is ideal for developers looking to build personalized AI assistants, customer support bots with extensive knowledge bases, or any application requiring an AI to remember and act upon user-provided information over time.