Telegram GPT Assistant n8n Workflow: Perplexity + Wikipedia

Turn your Telegram bot into a GPT assistant with Perplexity + Wikipedia answers (inside n8n)

This n8n workflow listens for new messages in Telegram and responds with an AI-powered assistant powered by an OpenAI chat model. It can optionally use Perplexity for real-time web search, Wikipedia for background info, and a calculator for arithmetic—while keeping short-term per-chat memory.

What this workflow does

• Triggers on Telegram messages: It starts whenever a new message is received by your Telegram bot.
• Loads your assistant configuration: System prompt, selected OpenAI model, optional Perplexity model, memory window, and maximum reply length.
• Answers via an AI agent: The user’s text is sent to an OpenAI-powered agent that can call:
  • Perplexity for web search
  • Wikipedia for reference-style lookup
  • Calculator for arithmetic
  • Per-chat memory to maintain short-term context
• Keeps replies within limits: If the generated answer exceeds your configured maximum length, it truncates the response.
• Returns the message to Telegram: Sends the final reply back to the same chat with web previews disabled.

Use cases

• Instant Q&A in Telegram: Answer “What is…?” and “Who was…?” using Wikipedia.
• Up-to-date research: Use Perplexity web search for current events or newly updated topics.
• Explain-and-compute: Combine factual lookup with quick calculations for estimates and conversions.

Technical details

• n8n nodes / tools used: Telegram, Set, Filter, Sticky Note, Perplexity tool, and n8n-nodes-langchainagent.
• Required credentials: Telegram bot token (used by Telegram Trigger + Send Reply), OpenAI API credential, and Perplexity API credential.
• Configurable behavior: System prompt, memory context window size, and maximum reply length.