After reading everyone's great comments in my last post, I'm going to move forward learning Claude Code now and not later. I realize I'm probably going to miss the boat on using Fable 5 but I'm going to spend the next two days learning how to use Claude Code and watching a few more videos on what people are doing with Fable. I honestly didn't think Fable would be something I would use anyway (I wasn't expecting to jump into Claude Code yet) but after seeing some of the things it's capable of, I'm thinking I need to move as fast as I can before it is taken away on Sunday. Or at least use it and learn it a bit before completely having to pay for it. I watched four of Nate's videos today and started his six-hour Claude Code course. I still have to do the things in the classroom but I'm kind of jumping to the head of the line learning Claude Code ASAP so I can at least try Fable. I don't know a damn thing about Claude Code, especially all the slash commands. Or using APIs, CLIs, and MCPs for that matter. But...I'm a firm believer in "Trial By Fire" so WATCH OUT WORLD - HERE I COME!!! 🔥🔥🔥 If any of you have any suggestions, I am one hundred percent open to anything you can tell me. 😀😀😀 Thank you!!!

I would like to share the progress on the AIS project. Although the recommendation is to use Claude, I have a Gemini account, so I used Gemini along with Opencode as the platform. The results are identical. I requested the AI to adapt the claude.md file, and it was successfully resolved. I am sharing the outcomes here. I understand that the majority of this group is English-speaking; however, as a Latino, I find the material shared here truly interesting.

If you build automations long enough, you hit a problem nobody warns you about: you end up with too many tools. I counted mine this week. 177 of them. Skills, agents, commands, plugins. Some live, some archived, some parked in reserve. At that scale the hard part stops being "can I build it" and becomes "which of these do I already have, and which is the right one for this job." You start losing more time rediscovering your own work than you ever spent building it. You rebuild things you forgot you had. You reach for the second-best tool because the good one slipped your mind. So I stopped trusting my memory and built an index. Here is how it actually works. One file is the source of truth. Every tool gets a single line in one register file: its name, what kind it is, its status (live, reserve, or archived), when I made it, and a one-line note on when to use it. Nothing lives only in my head. The dashboard builds itself from that file. A small script reads the register and renders the searchable page in the screenshot below. I never hand-edit the dashboard. Change the index, the page regenerates. One source, always in sync. Nothing is "done" until it is booked. My rule: a new tool is a booked tool. Building it is not finished until it has a row in the register and turns up in the search. That single rule is what stops the sprawl at the source. A drift check keeps me honest. A script compares what is actually installed against the register and flags anything that slipped: a tool on disk with no row, a row whose tool is gone, or an archived tool that quietly came back to life. Then I wire it into the day so it never rots: 1. Start of day. The drift check runs before I build anything. If the index and reality disagree, I hear about it first thing, and my usage stats refresh so the dashboard shows what I actually reached for lately. 2. End of day. The same check runs again as part of shutting down, and it blocks the wrap-up if anything is unregistered. Then everything backs up. The register cannot silently fall behind.

What skill I built: I initially took the easy road for my first skill and went with the content creator. My wife and I want to get back into consistent content creation, though, so this actually serves a greater need. I titled the skill nutrient-content, and I can either invoke it directly or tell Claude I want to create some content. I can either give it a topic and have it research the topic, or copy-paste a section of text or transcript and have it identify and synthesize the best takeaways for content with the explicit purpose of growing our audience. It will also format it either for X/twitter, an Instagram post or carousel, or an Instagram reel or YouTube short (I'll do the video generation when I have a paid Kie account, keeping it low-cost for now). One optimization I made: I included a few .docx files I have that go over content hooks, viral ideas, etc. I was pleased to see Claude improve the writing. But more so, I was even more impressed to see Claude tell me "a lot of this is generic engagement-bait ("You're NOT going to believe...", "This ONE trick will blow your mind", FOMO, Shock & Awe, tear-jerker emotional hooks). That directly conflicts with brand-voice.md's "no hype, no filler, no fear-based claims" rule, so I'm not adopting those wholesale." THAT'S what made me smile. I also ran into some initial issues with the image generation. Kie wasn't cooperating, and Canva was just downright bad. Finally got Kie working properly, and I'm happy with the initial results. I think I'm going to play around with Day 3 a little more. Two other skills I want to create over the next day or two: 1. a skill that pulls all the mechanistic data from all the functional health conversations Claude and I have, and compiles them into my pre-existing resources to supercharge the knowledge base. 2. a skill that I run weekly to round up and review recent research papers, detailing the key takeaways, coaching applications, personal applications, and content strategies.

One of the websites I scraped was a supplement brand I'm partnered with. Their product category is extensive, at 411 products by my count. So I did this to look at creating information for myself and all of our clients to quickly search for products based on use case and category, rather than slowly searching one product at a time. This would've literally taken me WEEKS to do manually, but it took like 2-3 minutes. MIND.BLOWN. One thing I learned was how easy and quick this process is. I always (somewhat foolishly) thought scraping was some dark arts level magic that only hackers knew how to do... Buuuut looks like I've got my Gandalf hat on now, buddy lol. I can think of a million other use cases for this that fit right into our business. Scraping equipment pages for home gyms Scraping websites for recipes for clients Scraping functional health websites for information on optimal blood lab ranges to use with our labwork interpretations that we offer... Needless to say, I'll be revisiting this project repeatedly and will be upgrading to a paid Firecrawl account in the very near future. This is powerful!