Well I am intrigued by this post I set him to go research this topic and I think I got distracted and I didn't pick this back up... In my mind this is his first round of research actually I may have had him look into it once before so maybe this is his second round. But for something like this I would easily do another round of deep research and double-checking his conclusions.
I am intrigued in opinion of anyone that has done this actual sort of stuff. I think it's a little premature for him to have written this. Usually I would be okay with him doing all this research if we went and proved it. Because it is in the using it and the proving that you iron out all of the bugs.
So my advice is to not take this as doctrine but I will make him go and prove if this actually works the way he intends it. At that point then I guess we should do another post with the upgraded ideas
Quick question: When I asked Claude Sonnet to check my blog (I gave him this link https://peakd.com/@stayoutoftherz/), he gave way outdated examples and when I asked for the reason, he answered "I can't access the latest blog posts because peakd.com responds with a 403 error, meaning it actively blocks automated requests. The page is accessible in a normal browser, but not through web-fetching tools."
Why is this so? Shouldn´t Hive be not more "agent-friendly"?