Does Synthetic Fabric dream of Active Inference (too)?

all right hello and welcome everyone this is active inference stream number 004.1 it's October 2nd 2024 and as we often do we'll start it out with a github push up went to three why not go with it this is getting pushed to the active inference institute github to the repo named fabric this is a fork of Daniel Miesler's collaborative project called fabric so let's verify that we got everything updated let's just double check that we're in and then we will okay that went to fabric just want to make sure that we got it going on the code here okay interestingly not there well we'll check in we'll check in later um this will be a short interesting stream of course open to anyone else's suggestions going to look at this fabric framework for orchestrating LLM and other kinds of synthetic interoperabilities and just look at a few of the functionalities that I've been playing around with today so it'll be rough it's 5 p.m but so it goes okay first I'm using cursor down on the bottom of the page you can see the change log I'm using cursor 0.41.3 um we're going to be looking at the fabric framework it took a little bit of playing around with golang and the installation environment to get fabric running but it's running it's a beauty it's fun to work with and we're going to be playing around with it and the output where this really gets one to is here's live stream let's say 57.2 here we have a quiz on active inference and real relevance realization we have the main idea being output we have the predictions testable predictions with confidence extraction of questions extraction of wisdom agents extract wisdom output improve academic writing the metadata for the youtube video summarizing the outputs of the whole video the transcript of the video a tweet a hacker one report and a micro essay and those are just a few of the patterns so let's pull back talk about the context talk about fabric jump into it and then also try to find out what is really happening with this what is happening maybe it wasn't pushed there we go there's the inference stream four okay it was there there was just a glitch in the matrix okay in the output folder that's where everything goes so let's uh jump in with seeing what it looks like when it's running and then we'll pop back to what fabric can do and then we'll play around with like a few next steps because there's so many cool things that can happen um let's look at the fabric documentation just briefly first okay so also for those who are watching live please write any comments or questions in the live chat it'll be hilarious and awesome to to see what people throw into the mix okay fabric is an open source framework for augmenting humans using ai there's introduction videos what and why since the start of 2023 and generative ai we've seen a massive number of ai applications for accomplishing tasks it's powerful but it's not easy to integrate this functionality into our lives in other words ai doesn't have a capabilities problem it has an integration problem fabric was created to address this by enabling everyone to granularly apply ai to everyday challenges philosophy ai isn't a thing it's a magnifier of a thing or process and that thing is human creativity so let's keep that in mind and think about the flourishing of the creativity and the learning as we continue on they talk about having too many prompts and this kind of trade-off between having a library of prompts and copying and pasting and manually doing that and do you sort that into folders do you make a database with tags how do you really get all those prompts or or do you meta prompt do you ask it to write a prompt that does this or that and then copy that or do you automate the asking of the meta prompt installation again it's going to be situational on your setup but it's a golang package so that's a new language for an inference stream and after getting the golang 1.23 installed getting fabric installed was pretty straightforward on my linux…