Bockit

I don’t agree. I think the way maybe the author used planning to maintain an understanding of the changes is largely not necessary, but there are other good use cases.

For example, it’s a great hook in the process for agentic review.

Get a second agent to look at what will be implemented and check it for inconsistencies, check it against whatever decisions were made or provided previously in the chat, or against whatever technical rules you’ve written out for your project, before going ahead. It surfaces a large number of opportunities for refinement, and generally pushes the output closer to the direction you’re looking for.

You can use the harness sdks instead of the model APIs directly. Eg if you use Claude agent sdk you still can have plan mode, questions, todo lists, sub agents, background tasks, workflows, nearly everything that Claude code has and taking advantage of how it + anthropic models are complementary.

I cannot agree more. We all have our own needs, and it’s never been easier to make something that meets them exactly.

I have been running my own setup and evolving it since November as I come up with more ideas and needs. The nicest bit is not being tied to one provider or another. Anthropic decides to stop supporting Claude agent sdk with subs? Or astra outstrips other models? My harness now works with Claude/cursor/codex and soon ACP. It didn’t start that way, it was originally all in on Claude, but now all the things I built on top of Claude now work with the others too.

Also, many energy retailers increased the daily 'connection service fee' to compensate. My dad's daily fee went up an extra 55c~, or +$200/year.

I think it's more: AI assisted engineering is a new skill people are trying to develop and we're on this collective experimentation process, working out how to use AI for engineering with varying degrees of success.

If that's true, any statements defining what is necessary to do should be ignored in that context. I'm still interested in hearing about what people tried, what results they think they saw, and then trying to apply those findings to my own processes.

Which is to say I don't think the pontificating is pointless, but as statements of Real Truth, I agree they're likely wrong. We're too early in the game.

I have a Mac mini and I put every device that needs access on Tailscale

It’s such a fun time to have 1+ decade(s) of experience in software. Knowing what simple and good are (for me), and being able to articulate it has let me create so much personal software for myself and my family. It has really felt like turning ideas into reality, about as fast as I can think of them or they can suggest them. And adding specific features, just for our needs. The latest one was a slack canvas replacement, as we moved from slack to self-hosted matrix + element but missed the multiplayer, persistent monthly notes file we used. Even getting matrix set up in the first place was a breeze.

$20/month with your provider of choice unlocks a lot.

Edit: the underlying point being, yes to the article. Either building upon the foundations of open source to making personal things, or just modifying a fork for my own needs.

I've been doing JS for nearly a couple decades now (both front and back) and I landed on the same approach a few years ago. Pick your absolutely minimal set of dependencies, and then just make what you need for everything else. Maybe counter-intuitive to some, I feel like I'm more comfortable maintaining a larger codebase with less people.

What's more, given the tools we have today, it fits really well with agentic engineering. It's even easier to create and understand a homegrown version of a dependency you may have used before.

This is probably a semantics problem. You’re right. The models don’t know how to mcp. The harness they run in does though (Claude code, Claude desktop, etc), and dynamically exposes mcp tools as tool calls.

What’s special about front-end developers here?

Taran's Sphere Scripting for Dummies was what taught me how to program as a 14-15 year old trying to do the same thing :)

Test post please ignore!

Essentially, you don’t.