n8n Telegram AI Support Bot: GPT-4 + Supabase RAG

Turn Your Telegram Support Bot into a Smart AI Assistant (Text + Files + Images + Audio)

This n8n workflow upgrades your Telegram bot into an AI-powered support assistant that understands text, documents, images, and audio, interprets user requests with OpenAI models, and replies with context-aware answers powered by Supabase RAG.

What this workflow does

• Multi-modal input handling: Accepts multiple message types—including voice/audio, documents, images, and standard text.
• Dynamic file interpretation: Processes uploaded content so the bot can extract and use the relevant information for the user’s question.
• RAG with Supabase vector database: Stores document embeddings in Supabase and retrieves the most relevant context before prompting OpenAI—enabling true Retrieval-Augmented Generation (RAG).
• More accurate support responses: Combines user input with vector-based knowledge to produce replies that are more relevant and human-like than basic keyword bots.

Use cases

• Customer support automation for businesses using Telegram, with answers grounded in your knowledge.
• Legal/Policy document Q&A and long FAQ support using embedded references.
• Product information retrieval from uploaded documentation and internal resources.
• Project documentation and long-form knowledge lookup for teams and creators building AI chatbots.
• Developer/analyst experimentation with AI + multi-format input + vector memory.

Technical details

• Knowledge base setup: Run the “Add to Supabase Vector DB” workflow manually to upload a document from Google Drive and embed it into your Supabase vector database.
• Core workflow components: if, set, code, switch, telegram, and supporting sticky note documentation.

Category: n8n automation workflow. Built for n8n users, automation engineers, and SaaS operators who want professional Telegram AI support with OpenAI + Supabase RAG.