mifydev

Kubernetes is the last thing I wanted to see recreated for agents. It’s like Multics of cloud, now for agents. Complexity for the sake of it, powered by your favourite YAML slop bowl.

I've got access to their sandbox, don't know if I should've, lol

Huh. The idea is interesting, but I don't understand the decision of keeping it chat-like: it looks like the mix of the worst parts of Google Docs and Slack, I struggle to see how this will help people track decisions.

This can be improved though - I think that simply a separate collaborative doc and a chat linked to that doc would work way better.

Is there an SDK? Can I create the containers programmatically?

Based on comments below the solution is to make the API public and publish the architecture design, along with the binaries.

This way the company can avoid spending too much money on open sourcing the code, and the community can just rewrite the server while keeping the original binaries running.

This is a very myopic and unnecessary cynical sentiment. It's not about you - agents just need to run without your computer being on all the time. Coding is a background task that needs to run unattended now.

I predict that ntsync will eventually evolve into full blown ntoskrnl.ko and there would be virtually no overhead on calling Windows API. You can almost call it a Linux Subsystem for Windows.

Is it just aligning to EB1A standards, or more rigorous and picky process in general?

Missed opportunity for a tongue twister:

Who coded the code Claude Code code?

I've been experimenting with similar concept myself. The linter loop is the only thing that can keep the agent sane in my opinion, and if anyone can generalize bun+tsc loop to other tasks, this would finally be a way to trust LLMs output.

I was annoyed at how Claude Code ignores my CLAUDE.md and skills, so I was looking for ways to expand type checking to them. So I wrote a wrapper on top of claude-agents-sdk that reads my CLAUDE.md and skills, and compiles them into rules - could be linter rules or custom checking scripts. Then it hooks up to all tools and runs the checks. The self improving part comes if some rule doesn't work: I run the tool with the session id in review mode, it proposes the fixes and improves the rule checkers. (not the md files) So it's kinda like vibe coding rules, definitely lowers the bar for me to maintain them. Repo: https://github.com/chebykinn/agent-ruler

Hm, speculating a bit, but it feels like NTSYNC is essentially a beginning of NT Subsystem for Linux, or maybe ntoskrnl as a kernel module. Feels like the most clean and fast way to port Windows, since the rest of the interfaces are in the user space in real Windows. Essentially should be almost without overhead: user: [gdi32.dll,user32.dll,kernel32.dll -> ntdll.dll] -> kernel: [ntoskrnl.ko]

nice, htmx is infiltrating into the browser standards

Pantheon is closer than we thought

You can just boot freedos to run them, it will execute in real mode which has the same cpu instructions as 40 years ago.

I'm quite concerned about x86 future, but the article has a point if you read it past the title.

It says that x86 is highly standardised - even with different combinations of chips, peripherals and motherboards you know it will work just fine. It's not the case for ARM systems - can you even have something similar to IBM PC with ARM?

I personally know that adding support for ARM devices on Linux is a huge and manual task - e.g. look at devicetree, it's a mess. There is no standard like ACPI for ARM devices, so even powering off the computer is a problem, everything is proprietary and custom.

I don't agree with the article though, x86 is dying and my worry is that ARM devices will bring an end to such an open platform like modern PCs are.

What about the screen refresh rate? Do they deliberately keep it at 60hz so people would buy a MacBook Pro?

Frankly, I don't like this kinds of takes. Yes, people are seeing more spam in their pull requests, but that's just what it is - spam that you need to learn how to filter. For regular engineers who can use AI, it's a blessing.

I'm a long time linux user - now I have more time to debug issues, submit them, and even do pull requests that I considered too time consuming in the past. I want and I can now spend more time on debugging Firefox issues that I see, instead of just dropping it.

I'm still learning to use AI well - and I don't want to submit unverified slop. It's my responsibility to provide a good PR. I'm creating my own projects to get the hang of my setup and very soon I can start contributing to existing projects. Maintainers on the other hand need to figure out how to pick good contributors on scale.

I've submitted it to web store, but I'm sure that the review will be very long. The extension requires a lot of permissions, with this kinds of things personally I'll trust more if i can build from source.

It is passing ACID1 test now!

That's a good point, I'll add support for other models shortly.

Backend service is definitely way to go if you want to serve models for the user.

So Bash and Edit tools are a bit weird, Bash tool is essentially JS execution, and Edit tool automatically generates a script that performs the edits on the page. These tools are needed for the model to explore the page, whatever it does at the end it creates a separate script that will be applied on the page load.

Yup, full access to DOM! Still needs a lot of optimizations, but the trick is that the agent reads the DOM as file, so it can grep parts of it naturally.

Oh that's cool! I've just used wxt to pack extension for firefox and chrome and just used typescript and plain anthropic api. My goal is to make this run fully inside the browser, without any helper binaries, like I've seen with others.

The agent is extremely basic now, there is no context tracking, so create new sessions often.

The problem with zsh, as with a lot of open source tools is bad defaults. If only zsh had a better out of the box experience then OMZ wouldn't be needed. I used maintain a lot of configs, over 10 years I just keep forgetting what did I set up. I don't want to be bothered with this anymore.

And yes, this version will directly push your changes to master, have fun!

Ivan here, looking forward to someone breaking the website :)

this is 100% true.

"complexity very, very bad" - https://grugbrain.dev/

that's one of the reasons why we open-sourced everything, we don't store the code, and you can verify that :) we're just running claude code on your codebase in an ephemeral container, only thing we save is your generated plan

please don't get confused, it's not wrapping oai at all! We're wrapping Claude Code here