I finally made the time to start the challenge ... and already have questions. The WAT CLAUDE.md mentions a workflow and a tool - workflows/scrape_website.md and - tools/scrape_single_site.py I got across it as I read through it because I don't only want to copy and paste, but really understand how this works. So where can I find those? I guess I can still continue without them, but if there is already something out there that helps me scraping websites, so I don't have to develop everything from scratch, that would be soooo nice 🤩 How did the rest of you solve this?

I actually watch the AIOS video to create my AIOS and i was wondering if we can use obsidian and Claude code without paying api credits to make his AIOS if someone can help me Thank you for your answer !

Hi Everyone, I had a quick question about PHI and how we can get around it when it comes to using claude and an IDE. I currently work for a medical company and there's so many ways to use AI, but my biggest concern is throwing it into Antigravity because of HIPAA. When I did some research, it seems like the only way you can get a Claude plan is if you have a big enough company and you do the enterprise plan. Does anybody have any insight on a way that they can either get around it or sign a BAA with Claude? I've seen apps that say you can use their system and go through a BAA with them, but I just find that sketchy. Any insight would be helpful. Thank you!

I've been building websites with AI and hosting all of my client websites on my own Netlify account. The problem is that Netlify has now moved away from the old free/legacy plan to a credit-based system, and it's becoming much more expensive than before. Since I'm still a beginner, I never charged my clients separately for hosting. Now I'm realizing that probably wasn't the best decision. I'm wondering how experienced developers and agencies handle this. - Do you host every client's website on your own Netlify account? - Or do you ask clients to create their own Netlify account and transfer the project to them? - Do you include hosting in your monthly retainer, or charge it separately? - If you're building Node.js/Next.js websites, what's your preferred hosting setup? - If you were in my position, would you start asking existing clients to cover hosting costs, or would you migrate them somewhere else? I'd love to hear how you structure this professionally because I want to build a scalable system instead of making the same mistake with future clients. Thanks in advance! 🙏

Hey everyone, I’m building a custom Node.js/TypeScript backend hosted on Render to receive and handle Instagram direct messages for my app. I am currently in Development Mode and hitting a very strange issue that is driving me crazy. Here is my setup: - Created a Meta Dev App (Customize Use Case with Instagram API). - Generated an Instagram Access Token with all required scopes approved (instagram_business_manage_messages, etc.) and verified it on the Token Debugger. - Successfully added an approved Instagram Tester account. - Confirmed that "Allow Access to Messages" is turned ON in the Instagram mobile app settings for the receiver account. The Problem: - When I trigger a dummy message via the Meta dashboard's "Test to Server" button, my Render backend successfully parses the payload and text content perfectly. - When my tester account reacts with an emoji (like 👍) to an existing message, my backend successfully receives the message_reactions webhook payload in real-time. - BUT, when the tester account sends a regular direct text message, absolutely nothing hits my server. No logs, no empty objects, just dead silence. It seems like Meta is explicitly filtering out text messages in Development Mode while allowing reactions to pass through. Is this a known restriction or a Catch-22 bug where text webhooks only fire once the app goes Live? Or am I missing some hidden permission? I really don't want to submit for App Review just to test standard text DMs. Any help, insights, or bypass workarounds would be highly appreciated! Thanks.