Hey everyone 👋 I’m currently building a personal AI system around my university workflow and wanted to get some opinions from people who’ve done similar things. Right now, I’m setting up an Obsidian vault where I store all my uni-related data (assignments, files, notes, requirements, etc.). The goal is to later use something like Claude Cowork as a kind of “AI assistant” that can: - handle assignments - help with bigger projects - support exam prep Since I don’t have API/admin access to my university system (Moodle), I can’t automate data collection directly. So I’m focusing on building a good input pipeline instead. My current idea is: 1. use a Telegram bot as a quick input (send files, notes, voice, etc.) 2. process everything through n8n (detect type, transcribe, etc.) 3. then store it in Obsidian, where AI can later work with it Main question: Do you think it’s worth building this whole automation layer (Telegram + n8n + auto-classification), or would you keep it simpler and just upload things manually and let something like Claude Cowork handle organizing + structuring the data? Second question: Has anyone found practical ways to work around restricted Moodle APIs? In my case: - API access requires admin permissions - so I can’t use HTTP requests to fetch assignments or course data Are there any realistic alternatives here? (e.g. plugins, integrations, scraping approaches, etc. — nothing shady, just curious what’s possible) Would really appreciate any thoughts or experiences @Ryan Nolan, @Matt Payne 🙌