AI Chatbot with Long-Term Memory & Telegram

This template sets up a fully functional AI-powered chatbot with long-term memory storage, note-taking, and Telegram integration. It leverages OpenAI/DeepSeek LLMs with n8n’s LangChain nodes to provide a personalized and context-aware chat experience. User messages are processed, memories and notes are stored in Google Docs, and responses are sent back via Telegram or directly within n8n.

The workflow is designed to create an assistant that not only responds intelligently but also remembers user details, preferences, and notes across conversations, ensuring meaningful long-term interactions.

Features

• 📩 Chat Trigger – Start conversations when a new chat message is received.
• 🧠 Contextual Memory – Short-term context handled by a buffer window; long-term memories stored in Google Docs.
• 🗒️ Note Storage – Save specific notes (reminders, instructions, standalone information) separately from general memories.
• 🔎 Memory Retrieval – Fetch past memories and notes to personalize responses.
• 🤖 AI-Powered Responses – Uses OpenAI (GPT-4o-mini) or DeepSeek-V3 models for conversational intelligence.
• 📚 AI Agent Orchestration – Custom role, rules, and system messages ensure helpful, context-aware, and user-centric responses.
• 📤 Telegram Integration (Optional) – Deliver responses directly to Telegram users.
• 📝 Google Docs Integration – Acts as the long-term storage backend for memories and notes.
• ⚡ Fallback Handling – Ensures natural responses even when no new task is given.