@Nate Herk I think this will also help you go viral because everyone is looking for something like this on forums. This will attract more new members to the community. I've been doing some brainstorming. It could be something like this: 1- n8n docs are sent to the Supabase vector repository. Also active modules, what they do, when they are used are documented and sent. (Tools like web scraper can be used for convenience) 2- The user's input in Telegram is vectorised and sent to the AI Agent. 3- It is determined whether this input is a new automation request or about existing automation. Different agents create a response according to the situation. 4- These agents are connected to the supabase vector store. With text splitter support. o3 is connected as a module. (When OpenAI releases Deepresearcher API in the near future, it should be connected to it) It might also be good to connect n8n forums in the HTTP module to the agent. From here it pulls the most upvoted Q&As related to the topic. I'm attaching a few files as an attachment, I hope it will help you. If doing something like this doesn't sound completely bad to you.