thatxliner

I didn’t make this lol; just something cool I’ve found

it's funny how the readme still says "written in Zig"

it's weird at the bottom

Because with so many metals in high entropy alloys, you can tune it to whatever, and that's why it's currently being investigated for potential room temperature superconductors.

Hello Hacker News! This is more of an experiment/statements I wanted to make about needing specs for things (e.g. for agents to output) more than an actual "spec for itineraries."

While schema.org exists, it is a massive specification with a lot of cases that are likely unused if you were to optimize it for application use. This project really just points at the tension between using such an extensive schema versus more "practical" and application-focused ones, especially when token costs are in the question.

Honestly, that's very cool

That was basically my intent with this project, but I took the laziest way to get there lol

If anyone is curious, I've been making this using DeepSeek v4 Flash with Claude Code as the harness

My first question would be where should I start reading? It seems like you modularized it into multiple assembly files (how does that even work?)

Good idea! I'm adding error messages + spans ala ariadne now

I'm wanting to read this repository as a learning tool, so it'd also be nice to include docs—even AI-generated docs, but obvious I'd prefer docs with your own design notes and decisions—about the architecture of the code.

Really cool project though!

You can also drop into direct Rust with the (rust "...") macro if I forgot to implement anything

For everyone who is shaming on the project for "not implementing enough," then you can definitely help me with it.

For everyone who is shaming on the project for being "LLM slop," sure but that's the reason why something like this can exist in the first place. The point isn't to be a finished, production-ready product. The point is to be an interesting work, and just a sly bit silly

I remember this group did something else a while back too.

In the midst of GitHub's outages and decline, many prominent people, such as Hashimoto are moving off of GitHub. Codeberg is one of the alternatives, and I've realized that we might want to make a CLI for AI agents similar to the purpose that `gh` has served for GitHub

So Pokemon Go but better

Didn't they write a whole blog post on why they chose Zig over Rust?

What about Gitea?

how do subagents work?

slightly unrelated but what’s with storing the entire codebase in the root directory instead of a nested src folder? It makes getting to the README a lot more difficult

Note that Rust is not in fact named after Fe2O3; it’s named after a resilient fungus of the same name

is there a CLI?

You have to try Codex. My friend's been trying to convert me for months and he was right all along: with Codex you don't GSD or whatever prompting metaframework. You rarely (I actually haven't need to do this at all) need to ask it to retry because its implementation is bugged: it literally just works first try.

Maybe that's because the harness maybe (not sure; haven't looked at their source code) has it baked in? Doesn't matter; the point is that it works.

Now, the one thing I heavily dislike is the UI it generates...it doesn't seem to realize that matching UI patterns with the existing codebase is quite important.

TY!!

Yea this actually has been going viral when I posted it on Instagram. Maybe I'll need to post this on HN in a different time or wait for someone else to organically post it :)

oops, I recently switched the storage to Git LFS and the deploy workflow didn't download the fonts

They’ve basically positioned themselves as a workplace app for years now. A fully integrated project management and documentation really is just asking for AI to be part of it

and I don't think they ever will unless they're highly competitive (hopefully that price they have stays? at least for users)

I was thinking of building this exact thing a year ago but my main stopper was economics: it would never make sense for someone to use the API, thus nobody can make money off of zero demand.

I guess we just have to look at how Uber and Airbnb bootstrapped themselves. Another issue with my original idea was that it was for compute in general, when the main, best use-case, is long(er)-running software like AI training (but I guess inference is long running enough).

But there already exist software out there that lets you rent out your GPU so...

I'm planning to set it up as a auto-marketing and user acquisition agent, but I'm also backpedaling on the idea since (1) people can probably easily tell that it's AI (2) that will produce negative reputation as a "slop company" (3) that's feeding into the dead internet theory

So in short, there's not much to do. I don't really have tasks I can just "hand off to OpenClaw"

You can export as plain text to a proper document formatting app. Quillium is designed for writing process and writing process only, similar to Scrivener’s philosophy.

looks a lot like vscode

I feel like people keep forgetting that it’s possible to code without ai, but yes arguably a lot slower, typically.