Welcome

MCP was not clicking for me, so the best way to learn was to build a thing with it. I have a plethora of notes everywhere and now I have a machine that drops rough ideas into my publishing backlog.

MCP is basicly an api for AI that tells agents how it's supposed to be used.

Building an app factory, that utilizes Cloudflare's free tier resources to have quick and dirty apps to demo.

When you can do whatever you want to do, what do you want to do? I'm personally not so sure, since most of the problems in my world have been solved. On the other hand, you don't know what you don't know.

If you can automate something away quickly, you should do it. Writing the article is the part that provides value, and scheduling provides a way to write about things so that you don't need to do it on demand.

Internet attention works in such a way, that it's good for you if you get traffic. Increasing the amount of attention grabbing you can do with no extra effort is therefore valuable.

As a builder, this is also a fun hobby project to have while the baby is sleeping.

Most of the effort was for the implementation was looking for the right buttons to obtain tokens on LI's page, which made me feel like a archeologist looking at ancient ways of work. We also implemented an email-notification system that notifies me when tokens to various integrations are about to go stale, and I need to go clicking around to obtain new ones. I may outsource that to traditional browser automation when I get the 1st email in about 40 days.

I needed to change email backend to Resend. I could not get it to work, even as I diligently pasted their console output to Gandi's DNS-records.