EPISODE 02

Omarchy Sunday Social: Demystifying Plugins

Sunday, September 20, 2026, 3pm EST60 min8+ speakers

Summary

as posted on X by @NixFred

Omarchy Sunday Social: Demystifying Plugins

Omarchy! Omarchy! Omarchy! X Spaces, Episode 2

Listen here: https://x.com/i/spaces/1wxWjloyOYnJQ

Recorded Sunday, September 20, 2026. 62 RSVPs, a room that kept filling, and the marketplace maintainer himself dropped in near the end. See you next Sunday, 3pm EST.

Last week was a free-for-all and it was fun. This week we picked a subject and stayed on it: what a plugin actually is, how you get started, how you iterate, and how you share it. Then we opened the floor and the room turned out to be full of people building things nobody had heard about yet. That second half is the good part.

WHO WAS IN THE ROOM

@ncfrontiersman Wes Grimes, co-host, built OmaStorm. 25 years a software developer, "got a little bored in the career, but man, Omarchy has just injected my veins with adrenaline." Me, co-host: Blip, Infomarchy, Burnbar. @hancore_linux, who runs the plugin marketplace and joined us for the first time. @m_tolhuijs, who built OmarKit because he watched @hancore_linux's issue list and thought we should help. @cfaulkingham, who has been fine tuning a local model for Omarchy and will tell you honestly how hard it is. @thetundeo in London, hosting the London meetup this Thursday. @MuchmoreIT, who designs high performance computers for the federal government and has been on a 5,000 GPU system and a 250,000 core supercomputer. @ZryMiller Zach, building a voice transcription and agent management app called Soto. @NoPowerPoints David, who published HyperTile Equalizer and GitHub Repo Tracker. Mihai, who got an Unreal Tournament engine running on Omarchy. Plus a room of people who listened, which counts.

PART ONE: WHAT WE TALKED ABOUT

Follow everybody

Before anything else. X used to have this nonsense about follower ratios. Throw that crap out. If somebody follows you and they have Omarchy in their title, follow them back. There is no ego here and the follower count does not matter. Send DMs, make connections. It costs you nothing.

Same house rules as last week. No politics. No egos. We did talk about the controversy going on, and I think that stayed inside the lines. As @ncfrontiersman put it: there is plenty of negative noise out there and plenty of other avenues for it. This is not going to be one of them.

What a plugin actually is, and the mental shift that has to happen first

Omarchy shifted away from "download the application somebody else made" to "customize your own bar." That is a technical change, but the real change is in your head.

My grandmother used to say, "You get what you get, you don't pitch a fit." For decades that was computing. The binaries were locked down and if you wished the software did one more thing, tough. A lot of newcomers are still standing in that mindset. They download plugins and comment underneath asking if you can make it do one small extra thing. I am not judging anybody for that, but I want to encourage them: do it yourself. You can. That is the whole point now.

I am not a programmer. Last week we talked about why agents give infrastructure people a bigger boost than they give programmers. 2X an infrastructure guy and it does not help much, I am already okay at that. What helps is being able to think of an idea and make it real.

And do not let the agent limit you to a language you happen to know. I tell mine: code it in whatever you think is best, Python, Rust, whatever. Drive the idea. Let it pick the tools.

@ncfrontiersman's version of the same point: when computers first showed up, plenty of people were scared to touch them in case they broke something. He never had that fear but watched people who did. With Omarchy and an agent there is even less reason for it. Please break it. Please touch it. Please mess with it.

Two first plugins, and neither of them was impressive

@ncfrontiersman liked the built in Omarchy Weather plugin but wanted sunrise, sunset and moon phase on it. So he opened his agent, cloned the plugin, and asked for it. Cloning is built in: the agent reads the Omarchy skill, copies the built in plugin into your own config, and it is yours to edit. Nothing is compiled, so it reloads in front of you while the agent works. "It's almost like building a web page."

The data was already there. Weather.in and Open Meteo were already being called and already returned sunrise and sunset. So the whole ask was: grab that from the same call, put it in a row under the forecast, give me an icon for each. All natural language. Then you iterate. He compared it to SimCity, building it up until it is yours.

Mine was worse. My first change was not even a plugin. I took the calendar in the center of the bar and added a small bar showing what percentage of the year we are through. That was it. Then I tweaked it to show two decimal places, and I thought, wow, it did it.

We are 71.94% through the year, by the way.

Here is why I tell that story. Even I was not thinking big enough, and I am reasonably good at the AI part. If you are learning Omarchy and AI at the same time, that is a wall. But the second one was bigger than the first, and the third bigger than that. Infomarchy started as something just for me and it was crap. Then about 15 people started contributing and finding things I had no idea about. @gh_TechLuddite has sent me stuff I read and thought, I do not know what he is talking about, but he is smart. There are still things out of my grasp. I lean on people for those.

Fork it. It is legal, and it is good manners to send something back.

There are no completely unique ideas. If a plugin is close to what you want, the license usually says you can fork it and make it your own, and that is the fastest path to something you actually like. When your agent finds a bug in the fork, send a PR back upstream. That is how this grows.

Full disclosure: I found @ncfrontiersman's plugin before I knew @ncfrontiersman, before any of this, and the first thing I did was fork it. I did DM him to ask if he minded, out of politeness, and he said go ahead.

One correction from the show. I credited the Trackpad plugin to "a guy named Fanta, I think it's David" and apologized in advance for getting it wrong. It is @davidfano. Sorry David, and thank you for the plugin.

The @dhh clip, which did not play

I tried to play a clip from @dhh's interview with @0xSero and the audio never made it into the room. My fault, I will do better next time. Here is what it said.

@dhh, on @ncfrontiersman: "He's been working on this thing called OmaStorm. So it tracks storms going over your area. And he put it into the weather widget that we have in Omarchy. It looks incredible. We're going to ship it in the next version. But here's a guy who just went like, well, I like storms. I want to see this lightning. I want to see where it's moving. He would not have done that before."

And the number in that same clip, already out of date when he said it: 3,000 plugins created for Omarchy in three weeks.

@ncfrontiersman's response to all of it was the most @ncfrontiersman thing possible: "It ain't about me. I would love to see everybody on this call have that opportunity." The barrier to entry is lower than it has ever been. Build the thing you personally love, share it, and see who else needed it.

The elephant in the room

A lot of the Arch Linux crowd is angry right now, and I want to talk about why without turning this into a fight.

Linux has been hard to use for a long time. For some people that difficulty became their identity. Running Arch was a badge of honor, ironically or not, because it meant you were willing to learn the hard stuff and break things and fix them. Omarchy changed that equation. The barrier is not your technical ability anymore. It is curiosity, persistence, and the desire to just do it. That opens the door to a completely different group.

Some of the pushback is the feeling that something small and yours is becoming available to everybody. From the inside, that reads as a loss. People handle loss two ways. They adapt and move forward, or they rebuild the wall around a smaller group. You can see both. Some look at Omarchy and think, good, more people are coming to Linux. Others are building a new badge of honor around not needing Omarchy. Now the cool kids run something harder, configure everything themselves, and can list all the reasons Omarchy is doing it wrong. Same wall, different garden.

There is a much more interesting move available to them. If Omarchy made the hard layer of Linux dramatically easier, then the people who liked hard problems should go find the next hard layer. Push into the kernel. Solve power management. Improve Wayland. Make gaming better. If difficulty is part of your identity, good news: there is an endless supply of difficult things left. Omarchy did not take away the frontier. It moved it forward.

@ncfrontiersman went further: he bets that if we actually got to talk to them, a lot would change their minds. People scan headlines, pick a side, and miss the nuance. He wants a big tent, because there are hard problems left and we need the people with the scar tissue, who have the muscle of sticking with something until it is solved. Those skills matter more now, not less. AI generating the code does not mean the thinking is done. You should never outsource your thinking to it.

And if they do not come over, that is fine too. You can lead a horse to water.

Build what you need. That is the whole strategy.

All of my stuff is for me. If I need it, I build it, and I hope when I put it out someone else needed it too. Occasionally it resonates. Burnbar 2.0 went out right before the Space for exactly that reason: I was not budgeting my subscription tokens well, so now it tells me when to stop for the day.

@ncfrontiersman said the same from the other direction: it goes better when you take something you built for yourself and cared about than when you try to work out which plugin would go viral. Focus on what you know. Somebody else likes the same stuff you do.

That is also where @dhh's point about capitalism lands. The old guard was openly anti-capitalist and then surprised nobody funded them. Build something good, and what matters to you will matter to somebody else.

Five ways to put a plugin out there

1. Keep it to yourself. Completely fine. I have about 30 forks that will never be published because they are only for me.
2. Post it on X and code in the open.
3. Host your own plugin site. I did that at omarchy.nixfred.com, mostly as disaster recovery and a single place to install everything on a new box.
4. Submit it to the official marketplace. You do not have to: anyone can point the omarchy plugin command at a GitHub URL and install directly. But it is where people discover things, and it is not much effort.
5. Facebook, I suppose. You will not find me there. I sold storage to them years ago, they pulled me aside and showed me everything they knew about me, and I deleted my account that afternoon.

What submitting actually looks like

Point your agent at the publishing docs and have it walk you through. You open an issue on the plugin repo, answer a few questions, and automated checks run against your code.

@ncfrontiersman's real example: OmaStorm has a quickshell popover plus a background engine. The scan came back and said the engine might download more files than needed and fill people's disks. Good feedback. He fixed it, edited the issue, they rescanned, and it passed.

One of the checks is the commit SHA, so the code they verified is the code on your main branch. To re-trigger a check you edit the issue and update it. @NoPowerPoints raised a real gotcha: sometimes the fix is editing the anchor point in the submission rather than pushing a fix and expecting it to get picked up. There is a sequence, and it is learned the hard way.

@ncfrontiersman's advice, and it is the big one: tell your agent at the very beginning that you intend to publish, and have it keep the requirements in mind while you build. Much cheaper than building a pile of things that will never be accepted.

And once people use your thing, you have customers whether you wanted them or not. Do not rip a feature out from under them. Do not change the config shape and break everybody. @ncfrontiersman built OmaStorm on an x86 VM, shipped an x86 only Rust binary, and heard from people it would not install on ARM64. Now he ships both. You find that out by shipping and listening.

Prior art, and not wasting tokens

I have been burned on this more than once. Partway into building something I stop and tell my agent: go see if somebody already did this better. It comes back with "15 people have tried this and 17 of them are better than you." Damn it. Tokens gone, time gone.

Do that first. Then fork the best one and make it yours. @ncfrontiersman called it by its real name: prior art. Before agents, the first step on any problem was finding out if somebody already solved it. That has not changed.

@ZryMiller added the part nobody says out loud: agents will gaslight you into thinking your idea is the most unique in the world. Push back. Ask it directly. "Oh, yep, there are 30 applications that already do this and they do it better."

@ZryMiller also asked the question none of us fully answered: when do you actually launch? He is building Soto on Windows, voice transcription plus agent management, close to T3 Code, and plans to port it to Omarchy in something more native than Electron because Omarchy people want things they can customize. His plan was to ship when he uses it daily without hitting constant bugs, and he is not sure that day comes.

My answer: keep the first feature set as small as you can stand. If you build all 15 features up front, you are playing whack a mole forever. He admitted he is already way too deep in features. That is most of us.

And treat the agent like a person you are training. Three minutes of instruction gets you three minutes of results. Spend a day teaching it what you want and it gets much better. I treat Larry like a coworker and told him to push back on me and be snarky about it.

Three tips before you submit

Do not submit on the day you build it. Even if you are happy with it, sleep on it for a week and you will find something to make better. I get my best ideas away from the machine, walking the dog.

Open it up to PRs. Let other people in. That is where the stuff you did not know about comes from.

Bring patience. @hancore_linux has a lot on his plate.

OmarKit: run the marketplace pipeline on your own machine first

@m_tolhuijs built OmarKit for exactly the problem everyone had been circling. He watched @hancore_linux's issue list and thought, we need to help this guy. So OmarKit runs the whole marketplace pipeline locally: the same tests, plus the things @m_tolhuijs reviews himself, because he knows what AI tends to get wrong.

He comes from a Laravel background and wanted test driven development for agent written code, something that can tell the AI flatly, no, this is not good. It has verify and inspect commands, and he is building a run command that wraps your plugin so the failure modes @hancore_linux checks for, memory leaks and similar, cannot happen in the first place.

His honest take on skills: they are genuinely nice and a lot of work has gone into them, but they only go so far. "I had so many countless moments where I said, did you completely ignore what I just said here." That is why OmarKit exists. It is released, somebody from the community has already tested it successfully, and he wants people to try it.

@ncfrontiersman has been working the same seam from another angle: a dev loop that symlinks your plugin into the plugins directory, installs it, runs it, and removes it when you kill the command, so you get web style hot reload while building. He offered to contribute it.

Two tools that already exist and deserve more attention

@cfaulkingham came in with the practical answer to everything we had just spent ten minutes wondering about.

There is an Omarchy security skill on GitHub. Search "Omarchy security skill" and it comes up. It is by @jankeesvw. @cfaulkingham used it on two of his plugins with great success, and it saved him the repeated cycle of CI throwing issues back at him.

Same author, another tool: OmaVM, which spins up Omarchy VMs on Omarchy so you can test your plugin in a clean machine. @cfaulkingham said those two helped him more than anything else.

He also did not plug his own plugins, which is very him, so I will: a couple are in the directory and the rest are on his GitHub.

The local model report from somebody who actually tried it

Last week the room talked about a small local model trained on Omarchy. @cfaulkingham went and did it, and came back with an honest result.

He started with a LoRA adapter aiming to replace the default Omarchy skill, and used OmaVM to test it. He found a LiquidAI model that runs acceptably on modest hardware, because as he put it, it has to run on a potato, and not many models do. Basic testing was not bad.

But it is harder than he expected. It overfits very easily. He automated the training runs with Grok deciding what to change each round, and you can automate nearly all of it, but without a lot of human input on the fine tuning data it will not be good enough to be worth it. He is looking for people to work on it with him.

@ncfrontiersman's read: there are so many man hours going into the frontier models that it is hard to compete on general capability. The interesting question is what a small model only has to be good at.

@thetundeo on how to start a meetup: just do it

I asked @thetundeo for the secret to the London meetup, since it has more people signed up than most of us could manage. He corrected me: it has not happened yet, it is Thursday. So @ncfrontiersman asked him prophetically what the secret is.

"Honestly, I just did it. I didn't overthink it."

He saw @dhh's post about the WhatsApp exchange with Toby around funding, pulled the car over, and called a friend. The friend had not even heard of it and said "Oma what?" @thetundeo told him, just trust me, give me a venue and I will sort out the rest. They gave him the venue.

He added that this is very unlike him. That is the part worth hearing.

He also asked @ncfrontiersman why he had been hunting for a good Rust repo, and the answer generalizes past Rust. @ncfrontiersman has 25 years of software behind him but never touched Rust, so his method is to find exemplary codebases that are not just pretty but performant and healthy, point the agent at them, and pull out the patterns and the places people hit pain. Then distill that into a style guide made of code examples.

His finding: writing a pile of rules into agent docs saying do this, do not do that, works worse than simply showing good examples. Show me a well structured controller, model and view. Do not describe them. Agents follow examples better than instructions.

@MuchmoreIT: ROCm 10.0, a Grace Blackwell box, and Omarchy Fort Knox

@MuchmoreIT has been a friend of mine for 20 years. I cannot say where he works, and that is legitimate rather than me being coy. He designs high performance computing and AI systems for the federal government, talks to Department of Energy people for a living, and has worked on a 5,000 GPU system and a supercomputer of roughly 250,000 CPU cores. He is relentlessly humble about all of it.

Three things he is working on:

He got ROCm 10.0 running on his AMD Omarchy machine, which took spoofing the device ID. Next he wants an Arch derivative underneath so it runs cleanly from an Omarchy perspective. His note for the AMD people: ROCm 10.0 brings massive performance improvements.

He picked up a Dell GB10, the Dell equivalent of an NVIDIA Spark: Grace Blackwell, ARM CPU paired with NVIDIA GPUs. It is on DGX OS right now, which is Ubuntu derived, and somebody has already done part of the work to get Omarchy on Grace Blackwell. He is going to benchmark one against the other.

And the long one: shortcuts for federal customers who would want Omarchy. That means a crypto validation suite tested at NIST, which he puts at the better part of a year, plus the hardening checklist that makes a system acceptable in a federal environment. @ncfrontiersman named it on the spot: Omarchy Fort Knox.

It is also the end of the federal fiscal season, so he is buried for the next few weeks.

@hancore_linux, who runs the marketplace, showed up

He joined because I asked him to, and I am glad he did.

He had spent the day on maintenance and necessary updates, and was upfront that he had not had much time for improvements this past week. Anyone watching the queue noticed.

The news: a new marketplace is coming, with a new submission system. @ryanrhughes is building the backend and @hancore_linux has been supporting on the frontend. He did not want to spoil it, since @ryanrhughes or @dhh will announce it properly, but he said it is taking shape.

On pre-validation his answer was honest. Use what exists: OmarKit, and the security skill. He cannot yet predict how the new system will check submissions. It will be agent checked, but how deep and what it requires is not settled.

Mihai got Unreal Tournament running

He reverse engineered an Unreal Tournament engine into working on Omarchy, and got it running on Omarchy Mac too, which was meaningfully different from the x86 side. It runs well on the current kernel. We lost his connection before we could ask anything else, which is a shame, because that is a hell of a thing to do on a Sunday.

PART TWO: THE QUESTIONS STILL UNANSWERED

Last week we ended with twelve open questions. Several got answered this week, which is the point. These are the ones still standing.

Q1. How do you know what the marketplace will check before you submit?

This came up three separate times from three people who had never spoken to each other. @NoPowerPoints asked it directly. @m_tolhuijs built OmarKit to solve it. @cfaulkingham found jankeesvw's security skill and OmaVM. @ncfrontiersman points his agent at the marketplace source to work out the checks ahead of time. @hancore_linux says a new submission system is coming but cannot say what it will require yet. Three partial answers, no official one. Does a preflight skill ship with the new marketplace, or does the community keep owning this?

Q2. Is it safe to let an agent do the submission for you?

@NoPowerPoints said the quiet part: having the agent run the initial submission is great, and also a little scary, because it is using your credentials. Nobody in the room had a policy for that. What should an agent be allowed to do with your GitHub credentials, and what should always be a human pressing the button?

Q3. When is a thing ready to ship?

@ZryMiller asked and we gave him tactics, keep the feature set small, but not an answer. He is waiting for the day he uses it without hitting constant bugs and suspects it never comes. What is the actual signal that a thing is ready to put in front of people?

Q4. What does a small local Omarchy model actually need to be good at?

@cfaulkingham did the work and reported back: overfits easily, has to run on a potato, automatable but weak without serious human curated training data. Competing on general capability is a losing fight. So what is the narrow job where a tiny local model beats sending the prompt to the cloud, and who builds the training set?

Q5. With 3,000 plugins in three weeks, how do you find prior art before you burn the tokens?

Everybody now has the habit of asking the agent if somebody already built it. That is a workaround, not a system. Discovery across the marketplace, GitHub and X is still manual.

Q6. Does the big tent actually work?

@ncfrontiersman thinks a lot of the Arch crowd would change their minds if we got to talk to them. Nobody tested that this week. Is there a real move here, or do the walls just relocate?

Q7. Fleet management and rebuilding a box from scratch.

I raised it and we moved on. I run my own plugin site partly as disaster recovery, and we put import and export into Atmos as a piece of it. The Omarchy crew will probably beat me to the full answer, but right now there is not one.

Q8. What happens to a community plugin when it goes into core?

OmaStorm is going into the next version of Omarchy, the first time that has happened to somebody in this room. What changes for the maintainer, for the people sending PRs, and for the people who forked it? We did not ask @ncfrontiersman directly and we should have.

Q9. Who does the federal hardening work with @MuchmoreIT?

A NIST tested crypto validation suite is the better part of a year for one busy person. Omarchy Fort Knox is a real opportunity and right now it is one guy with a day job at the end of fiscal season.

ASKS ON THE TABLE

@m_tolhuijs wants people to try OmarKit and reach out. He is building it as we speak, aimed squarely at the submission pain.

@cfaulkingham wants collaborators on the local fine tune. He has a LoRA adapter, a model that runs on modest hardware, and a clear view of why it is hard.

@ncfrontiersman wants to see a proper plugin dev loop with hot reload, and offered to contribute his version to OmarKit.

@hancore_linux wants patience while the new marketplace gets built, and recommends OmarKit and the security skill in the meantime.

@MuchmoreIT is building toward federal ready Omarchy and would welcome anyone who knows that world.

@thetundeo's London meetup is Thursday. If you are anywhere near, go.

Everybody: follow everybody in this room. Ratios are nonsense. Send DMs. Make the connections.

NEXT EPISODE

Same time next Sunday, 3pm EST. Same format: come tell us what you are building and where you are stuck.

If you have never built a plugin, here is your homework, and it is smaller than you think. Clone a built in plugin with your agent and change one thing about it. One. Mine was a percentage bar on a calendar and it changed how I think about computers.

Do not just accept things as they are. If you can think of it, build it. And when you see somebody struggling in the comments, help them. Get their dictation working and they are off to the races.

Thanks to @ncfrontiersman for co-hosting, to @hancore_linux for the marketplace and for dropping in, and to everybody who spoke.

As @ncfrontiersman closed it: everybody's a special guest.

See you next Sunday.

Transcript

machine generated · speakers are their own X handles

All right, happy Sunday, everybody. We're going to get started here in just a minute. I'll give people about 5 more minutes to trickle in.

We had about 62 people RSVP, so we've just got, what, 6 or 7 in right now?

Those are the best ones, but...

Go ahead and come in.

Those are the best ones to get here early.

Yeah. They get the jam to the grooves while we're waiting.

What's the weather like out there, Wes?

Today it's muggy, a little cloudy, but I saw where it's going to be like fall next week coming up, so...

Wes, I handed you a softball draw.

I could talk about weather for hours.

We'll talk about it. I can't wait.

All right. Yeah, so what about you, Fred? How's the weather doing over...

It's hot, man.

... in your neck of the woods?

I'm in Atlanta near Chastain Park, if you know that area, and it's hot. The dog wants to go out, and I'm not doing it in 90-degree weather.

Actually, it's 85, and I know that because I have a Marshy Weather run.

There you go. There you go. Yeah, yeah, that 85 don't tell you the full story.

You need to know the dew point because that's what makes it, you know. If it's 85 and it's, you know, low dew point, low humidity, it's not too bad, but when it creeps up...

Exciting stuff, right, folks?

This is going to be great. I'm excited.

Yeah, I am too. I think it's going to be really good. Second time doing it.

And we're going to be talking a little bit... last week was kind of a free-for-all, and it was fun, but this time we're going to kind of stay a little more focused and talk about what a plugin is, how you can get started, how you can build and iterate on it. And at the end, we'll give folks time to answer, you know, to ask some questions and discuss a little bit, so.

All right.

Hey, while we're getting started, those of us that are here on time and early, go ahead and follow everybody. That only makes the community stronger. You know, Twitter and X used to have this bullshit about, you know, your ratios of follows, the how much... throw that crap out.

This is a new paradigm. If somebody's following you and it has a Marchi in their title, follow them back.

Like, there's no ego here. The follower count, who gives a crap?

Yeah, yeah.

And I see all the hearts out there, so follow your brothers and sisters out there. Make connections. Send DMs.

Let's change the way this thing interacts.

100%. Totally agree, Fred.

Yep. Also, you know, and as we're doing housekeeping, you know, Wes, I just want to talk a little bit about the same rules as last week. You know, no politics here.

We may talk a little bit about the controversy going on, and I think that's within the boundaries of no politics. But no egos.

You know, as I said, follow everybody back in the room. We're small, and we all have the same likes, so follow everybody back.

And I'll shut up, Wes. Take it.

No, yeah, yeah. Exactly. Yeah.

Keep it positive. Let's talk about all the good going.

There's plenty of negative noise, right? And so it's like, I don't want this to be another avenue of negativity, you know?

I mean, there's plenty of other avenues for that.

All right. And also, if you guys don't mind, if you could just tweet out that you're listening, and let's see if we can build this thing up. And I can't wait to hear from you right now.

And yes, I'm speaking to you right now. I can't wait to hear your story.

All right. Well, I guess, you know, we're coming up on 5 minutes here. We can probably go ahead and get started.

Fred, do you want to introduce yourself?

You bet. You bet. I'm Fred Nix on X.

See what I did there? Happy co-host with Wes here.

Wes is the guy that built Omastorm, which if you watch the DHH interview this week, you're going to find out that it's headed to the next version of Marchi itself. And Wes, we're going to get to that.

I know you're a humble guy, and you don't want to bring it up, but we're going to talk about it. Last week, we had about 50 folks, and we're hoping to build it up again.

We already talked about the house rules, no politics. And just a quick weekend and review.

Marchi, we shipped our own kernel. DHH said Quatro passed 300,000 downloads in the month.

The marketplace added 63 plugins in one day. The Mac fork got Touch ID working on Silicon.

Now, Wes, I know that's near and dear to your heart. Startup meetings are the meetups this week.

We're in London, England, Toronto, Copenhagen, and Atlanta. So let's get started, Wes.

Back over to you.

Yeah, wow. Yeah, it's been quite a week, for sure. Yeah, I'm Wes Grimes, North NC Frontiersman on Twitter, or X, as it is these days.

And yeah, I'm kind of new to the Omarchy community. I've been a software developer for about 25 years.

Got a little bored in the career, but man, Omarchy has just injected my veins with adrenaline. You know, it's just so much fun.

And the community has been great. Yeah, so today we're going to talk about demystifying plugins.

What is a plugin, you know, and all that. So I'm going to hand it over to Fred so he can kind of walk us through that, and then we'll kind of dive into, you know, how you get started, like building your first plugin.

Yeah, and I see a lot of the folks out there are experts and probably know more about the plugin stuff than I do. So, you know, chime in if you got a better explanation than I do. Plugin is a change a little bit from how software runs.

I know Marchi kind of shifted away from applications to customizing your own bar. And I've made a handful of them, and they've had some success.

And I look out there and could name names, Tech Lite being one of them, that's really helped me move forward. And I don't claim to be a programmer.

If you were here last week, we talked about the advantage that we believe that AI agents bring to infrastructure folks more so than programmers. If I were to be a 2X infrastructure guy, you know, I do okay at it, so 2Xing it really doesn't help me.

But giving me the ability to just think of an idea and make it, you know, come to life in... oh, shit. Oh, sorry, my laptop blanked out.

I thought I was going. So the agents give me the ability, if I can think of it, you know, I can make it happen.

And I see so many times in the plugins that I release, I get folks that say, that comment below, "Can you make it do something very simple," whatever that is. And this isn't me judging them at all, but they'll say, "Can you do something very simple?"

And I want to encourage them, like, I don't want to sound like a jerk, but kind of do it yourself a little bit. Like, there's a mindset of the past computers where we were kind of locked in, and you kind of got what you got.

You didn't, you know, my grandmother used to say, "You get what you get, you don't pitch a fit." And for decades, we've been in this paradigm where you kind of got the software that you got because the binaries were locked down.

And it's taken time for people to understand that. I think the newcomers to Omarchy, you're still in that mindset of downloading applications.

Somebody made it, and I can use it. And boy, I wish it did this, but it doesn't.

And they haven't shifted that mental model to, "I can just do it myself." So that's where the plugins for me are a big deal because my mind has switched to, "I can't do that," to, you know, anything I can dream of.

Now, am I a kernel guy? I mean, I could poke around in a kernel, but I don't even understand it well enough to tell it what I think I need the kernel to do at a higher... I can do it at a higher level.

But the kernel folks that get down into the details, you know, the key is to have the agent drive your idea and not limit you to a certain programming language or, you know, I always ask it, you know, "Code it in the best thing you think is possible, be it Python or Rust or whatever it is."

So the main thing, and then I'm going to hand it back over to you, Wes, is the main thing is don't just accept things as they are, right? If you can think about something that you want or need, do it, man. Just do it.

And help somebody else to understand that too. Whenever you're in the comments, you see people, you know, saying things, "I can't see all of X," and none of us can.

But if you're seeing people struggle, like, tell them, just talk to it. Help them get dictation working, and then they're off to the races.

What do you think, Wes?

Yeah, no, I think that's right. I love it. I mean, you know, when we first got around to computers, a lot of people had a fear, like, "I'm scared to touch it.

I'm scared to break it," and all that kind of... I'm one of the people that never had that fear, but I knew a lot of people around me did. And I think even more so today, with, you know, agents and Omarchy and the way it's a malleable operating system, it's even less of a concern.

Like, please break it, please touch it, please mess with it. You know, have fun with it.

Create, you know, if it's an idea you've got and you have that agent, you can go at it. And so, like, how do you get started?

Like, I think honestly, like, maybe the first level of, you know, customization is like just the OS and configuring. Yeah, hey, I want to have rounded corners in my windows, but that's pretty cool.

You can do that with Omarchy. But like, you can actually start to build what you want to see.

And so for me, for an example, like I started out, I liked the Omarchy Weather plugin, but I wanted to see like the sunset and the sunrise and the moon phase. So I opened up my agent, and I think at the time it was Codex, and I was like, "Hey, what can we do here to add those?"

And so the cool thing is kind of built in is this idea of cloning a plugin. And so you can tell your agent, it reads in the Omarchy skill.

It's like, "Okay, I'm going to clone this Omarchy built-in plugin." It makes a copy of it into your config that's your own that you can edit and change, right?

And because it's not compiled, like, as your agent's making the changes, you're making it, you can kind of see it reload right before your eyes. It's almost like building a web page.

It is. It is.

Yeah. And so, you know, it's like, you say, "Hey," you know, you start kind of thinking about it. Where is this information I want to show available?

Is it being fetched already? And in this case, it was using weather.in and also OpenMateo.

And those services already provided the sunrise, sunset info. And it was really as simple as just saying, "Hey, can you also grab that information from the same call and show it on the screen?"

And I'd like to see it as a row below the forecast row. You know, show me a little icon for the sunset and the sunrise and the moon phase.

And this is all I'm doing. I'm just talking in natural language, explaining it, you know?

And then, and before, and you kind of, it's like this, it's like, I don't know if y'all used to play, you know, like SimCity or something. It's like you're building this thing over time.

You're iterating over it, making it yours, and it becomes a lot of fun. And you get to a point, and it's like, "All right, I love this plugin.

I've used it for myself. Maybe I want to share this with other people."

And so that's something we can talk about in a minute. But, you know, always the start is like, you don't have to build it from scratch.

You know, that always is hard when you're having to figure out, "Oh, I don't know the shape of a plugin. I don't know what right files I need to have."

So like, if there's a plugin that's sort of doing what you want, you can have the agent clone it, and you can start tweaking it to your liking. And then, you know, it's kind of all up from there.

Yeah, so Fred, what about you, man? Or anything you want to ask me first?

Yeah, so, well, I want to add on a little bit what you said about the forking. So, you know, there's no perfect or there's no completely unique ideas out there. And I like to be careful.

A lot of folks, I look at plugins and I think, "Well," and I'll start using it. And like the Trackpad one, for example, which was made by a guy named Fanta, I believe.

I think it's David, and I should know it better than that. I apologize if you're wrong.

So the license and the way that open source works is that it's 100% legal and okay to go take that fork and make it your own. And once you do that, it's, you know, kind of a good thing to send back a PR if your agent kind of finds bugs.

And so let me transition to one thing real fast is that I found Wes' plug before I became friends with him and before I followed him and before I, you know, before we started talking. And the first thing I did was I did fork it.

And Wes, I think I actually emailed or DM'd you and said, "Do you mind if I fork this?" You know, out of politeness.

Yeah, yeah.

And he said, "Certainly." But, you know, Wes has come a long ways with his Omastorm. And this week, DHH called him out specifically on a podcast with ZeroXSero.

I'm not exactly sure how to say that, but his at is @zeroxsero. Very intelligent young man.

And I just want to play a little clip. Let me know immediately if you can hear it okay about DHH in the interview talking about Wes' plugin.

Here we go. Track two.

Guys, a tweet I just saw this morning. He's been working on this thing called Omastorm. So it tracks storms going over your area.

And he put it into the weather widget that we have in Omarchy. It looks incredible.

We're going to ship it in the next version. But here's a guy who just went like, "Well, I like storms.

I want to see the storms. I want to see this lightning.

I want to see where it's moving." He would not have done that before.

We have, actually, this is outdated now. A week ago, it was 3,000 plugins created for Omarchy in three weeks.

Once you've lived that, working with other people and trying to coordinate.

Oh, you're not hearing it?

Oh, all right. Well, I'll do better next time.

That's okay.

And I'll reload. Sorry about that, everybody.

That's all good.

I'll post it in the chat.

Yeah, yeah, y'all can watch it. We'll figure out how to do that next time. But it was cool.

Somebody shared it first with me and was like, "Hey, did you, maybe it was Tundy that did." I can't remember, but.

Tundy, my man.

He's already started a, I believe he's doing the London meetup and doing amazing, like doing, it's blowing up. So good on you, Tundy.

Yeah, yeah. And I think what DHH was really just saying is like, it's a cool platform. And if you have a good idea and something that you love personally, get out there and make it, you know, and share it.

And that's really kind of the point there. So.

Fantastic.

Yeah, it was a neat little surprise to wake up to. But it ain't about me. I mean, I would love to see everybody on this call, you know, have that opportunity.

So I think it's just a lot of fun when your work is shared and people find value in it. And this is like the barrier to entry here is so much lower than it used to be.

Oh, man. Oh, man. Hey, Wes, you mind if we kind of switch topics a little bit?

So I want to talk about the elephant in the room this week. And that is all of the Arch Linux crowd that's very angry about their own walled garden of sorts coming down. And Linux has been tough to use for quite a long time.

And for some people, it kind of became their identity. Running Arch, ironically, was kind of a badge of honor because it meant you were willing to learn the hard stuff and break stuff.

But Omarchy really changed that equation. And the barrier to entry is lowering.

And whether you already know Linux well enough to survive Arch install, the barrier isn't any longer your technology ability, but it's really become curiosity, persistence, and the desire to just freaking do it. So that opens the door to a completely different set of group.

And some of those Arch Linux guys that are hating on Omarchy, some of that pushback comes from the feeling that something that they would belong to a small club is becoming really accessible to everyone. And from their perspective, it's kind of a loss.

And people tend to react to loss in one of two ways. They adapt and move forward, or they find a reason to rebuild those walls around a smaller group.

And I think you can see a little bit of both of that happening because some of those guys see Omarchy and think, "Awesome, more people are coming to Linux." But then others are creating a new badge of honor around not needing Omarchy.

And the club becomes smaller again. And now the cool kids are the ones who know better, that run something harder, that configure everything themselves, or can tell you all the reasons that Omarchy is crap and doing it all wrong.

And so we're talking about the same wall again, but a different garden. But there's a much more interesting move available to those people.

If Omarchy made that layer of Linux that was difficult dramatically easier, then those smart people should go push and find the next hardest layer. Maybe push into the kernel, maybe solve power management, prove Wayland.

Let's make gaming better. And if being difficult was part of their identity, great.

It's almost like an endless supply of difficult things to fix. So Omarchy didn't take away the frontier.

It just moved it forward. So what are your thoughts on that, Wes?

Yeah, no, I 100% agree. I think, number one, I bet if we got a chance to chat with them and show them, they might actually change their minds. Because a lot of times people just kind of read scanned headlines and, you know, they jump into their sides or whatever.

And they miss the nuance. But I would love to see every single one of those Arch people, you know, come on in and, you know, let's just make a big tent, right?

Let's see what we can do to, because there are so many hard challenges yet to be solved. And we need people that have built up that scar tissue.

They've built up those muscles of sticking out a problem that's really hard until it's solved. And I think that those skills apply even more now than ever.

You know, even though AI can generate the code, that doesn't mean it's doing everything. You should never outsource your thinking to it, right?

Like, it's just a, it's just kind of a levels the playing field in some ways, but again, you still need to have that agency. And so I would love to see those folks, you know, come over.

And if not, that's cool too, whatever, you know?

Yeah, you can lead a horse to water kind of thing. But at the end, as DHH says over and over, we are going to win. And we're going to beat, you know, trillion-dollar companies.

And, you know, it's just amazing that all of these things have kind of come together. Any one of them alone, you know, AI by itself wasn't doing it.

Omarchy before AI wasn't doing it.

But, you know, the quality of people that I've met, some of the people that I know that I've known for 20 years that are on this chat right now, you know, it's just amazing. Wes, what's next on our agenda, man?

Yeah, well, before we hop into building the interview, I wanted to hear a little bit about your first plugin and how that went.

All right. So that's a really good point. And, you know, earlier I was saying that you want to talk about, talk to people in the comments about, you know, all you got to do is just think it and dream it.

And I was just thinking, when I first installed Omarchy, not to be an asshole, but I'm pretty good at the AI stuff, you know, more than if I go to Publix and go to the grocery store. Likely, I'm probably the best AI guy in the grocery store.

I'm not egotistical, but there are many people on this chat right now that I know by name that are better than me. However, I'm okay at it.

So when I first got Omarchy and I thought, "Okay, I can make changes," even I, as an AI understander, still wasn't thinking big enough. So the jump for somebody that's doing both, that they're learning Omarchy and learning AI is a big wall.

Because to your question, what was my first plugin? My very first thing, and you're going to laugh at me, was I took the calendar date, the one that at the very center that we all have, and I added a graph bar that told me what percentage of the year we are through.

And that was it.

So just for your knowledge, we are 71.94% through the year.

And I thought, "Wow, that's crazy." And even that blew me away. But the crazy thing is, is that once you understand that anything's possible, from there, my next plugin was bigger.

And from there, my next plugin was bigger.

And from there, you know, Infomarchy started out as an idea just for me, and it was crap. And then some 15 people have started contributing and finding stuff, again, that I had no idea about. And getting pretty technical in the weeds, like Tech Luddite sent over some stuff that I read.

And I'm like, "I don't know what the hell he's talking about, but he's smart." So, you know, and don't take that as I'm an, I guess I am an idiot for just accepting it.

But, you know, there's things that are out of my grasp that are still out of my grasp. And I rely on the people that I know and love and getting the love to help me.

So my first plugin was not even a plugin. It was just to edit the calendar that's at the top of your bar that shows the date to say what percentage of the year we're through.

And we're through 71.95% of the year. And the tweak that I did was I told it to do two decimal points.

And I was like, "Wow, it did it."

So take a step back. So you've installed Omarchy. You wanted to add, you came up with the idea.

What did you do? Open up terminal, pull up what agent did you use?

And did you say, "Hey, clone this plugin?" Or how did that?

I've kind of, before Omarchy, you know, and I would look to build things, I built a bunch of stuff. And a lot of my stuff is infrastructure. And you can go to nixfred.com and see all the crap that I've built.

And so that was where my focus was. But I didn't really have an operating system.

And I was doing it for a Mac. So I was kind of held behind, you know, the Mac, you know, there's a lot of talk about agents really aren't that good on a Mac.

And they're kind of not. So when Omarchy came out, I kind of, it took me a couple of days to kind of understand what it was.

And I was doing things for my own benefit and things that I liked. And then, you know, Atmos came out by a good friend of mine.

And I was able to submit some PRs to that. And then I just started helping other people with their things.

And I've been accused of having a little bit of an eye for visualizations. And what I found is that if I like it, somebody else does.

And I think the second thing I did was tell it to, and don't hate me for this, but make it look more like a Mac and make the Alt Tab more like a Mac. And then pretty early in, I said, "If you can just show me what's on my iPhone for iMessages, just like if I could just see it, that'd be really cool."

And I was very happy that I was able to see it. And then it grew from there.

And so BurnBar, which I just released 2.0 out, if you want to check it out, it's a big, big improvement in my opinion. But all my stuff is for me.

If I need it, I build it. And I hope that when I put it out, that other people need it as well.

And it turns out that occasionally it resonates.

Yeah, yeah, totally. Totally. Yeah, because I mean, it just seems to go better if you end up taking something you've built for yourself that you love, that you take time and care in building, then trying to figure out, "I wonder what plugin would go viral."

Or, "I wonder what plugin everybody could use." Like, just focus on what you know.

And then I guarantee there's going to be other people out there that like the same stuff and would benefit from you sharing, you know, what you're doing. So.

Exactly. And DHH talks a lot about how we've got to get capitalism involved and how the old Linux guard was anti-capitalism. And then they were surprised that nobody would give them money because they were directly a threat to what, you know, Linux was doing.

Why would you give money to people that, you know, their webpage says down with capitalism? So what happened, you know, what DHH says is build what you think is good for you.

And, you know, the things that matter to you probably matter to somebody else.

Yeah, yeah, for sure. All right, so let's fast forward. So talking about this idea of we've come up with an idea, we've made a plugin, now we want to share it.

Now we want to get it, you know, we want to get it out there for other people to try. Like, to start with, you don't have to get it published in the plugin directory. That's a great place for people to discover and find it.

Like, it's not one of those things like with Apple, right? Or, you know, other operating systems where you have to have it in the store and get approved.

Like, anybody can take the command, Omarchy plugin command, point it at a GitHub URL and install it, which is pretty cool.

Yeah, and so what I used to do was, you know, I don't know, I'm sure raise your hand or clap or whatever. If you kind of, before Omarchy, we're running out of things to build on AI. And that kind of sounds crazy.

And, you know, we're just looking for that next dopamine hit. You know, AI kind of has replaced the doom scrolling for the folks that get onto it.

So it became a thing for me to fork things and then make it my own.

And so I kind of had that natural thought process. And so it really, all the things it feels like in the past year and a half with Claude code and seeing it from before it was publicly released all the way to where it is now at 2.1.200 and something, I think it is, you know, we've come a long ways.

And the tools are getting better. And so, you know, I certainly encourage people to fork things.

There's probably 30 forks that I've made that there's no reason to publish because it's just for me. I am a good citizen.

And I send back a PR if I think it might help the overall project. But, you know, forking things is a great way to do it.

And you're talking about how do you get it out there? There's really five different things you can do.

Number one is you just keep it to yourself. And that's fine if it works for you.

Number two is you can go to X and you can, you know, start publishing out there and code out in the open and let other people see it. The third thing you can do is you can build your own website and kind of host your own plugin library.

And I've done that. But really, that was at Omarchy.nixfred.com.

The only reason that I did that was really to, as a disaster recovery for me and just kind of a one place I can go to install things for a new box. And I'm also working on a way to, I think the Omarchy crew is going to beat me to it.

But there's some things in the way of fleet management and being able to recreate boxes on the fly. We added a piece of that import-export to Atmos, if you've seen that.

But so we got keep it to yourself. We got show it on X.

We've got make your own website that publishes your own plugins. The fourth one being, you know, advertise it and submit it to the official plugin site by hackers.

And, you know, he does an amazing job. And the fifth thing is, and I don't know if this is even popular, put it on Facebook or whatever.

But for those of you, if you're looking for me on Facebook, you're not going to find anything. I went to Facebook headquarters and sold them some storage way back in the day.

And they pulled me aside and said, "What's your user account here?" And I told them.

And they told me and showed me all the things that they know about me. And I deleted it that afternoon.

Ooh, yeah. Yeah, that's always scary. Yeah, so yeah, so you don't have to share it if you don't want to.

You can share it on X. Like, you don't have to publish it to the plugin directory.

But like, let's say you come up with a cool idea and you do want to share it. Like, the idea is there's really good docs, right?

Like, you point your agent at the plugin, you know, how to build a plugin, how to publish a plugin, and say, "Hey, step me through doing this the right way." And there's going to be a process where you open an issue up on the plugin repo and you have to fill out a couple of questions.

Well, you don't have to. The agent can do it.

And then there are some things that get validated. So like, for example, for me, with Omastorm, my plugin has the quick shell, like pop over window, but it also has this engine that runs in the background.

And so I submitted it and actually appreciate it. They did the scan and they said, "Well, your engine might accidentally download more files than needed and cause like people's disks to fill up."

And so it was really good feedback. And so I went back and I made that fix.

And then I, you know, you edit the issue. And then they do another scan and come back and say, "All right, yeah, I see you fixed that.

It looks good." And so there's, someone has created a skill, I think, that makes this a little easier.

But if you think about this stuff at the beginning when you're building it and you tell your agent, "Hey, I'm going to be publishing this at some point. Let's keep the requirements in mind," you know, as we do it, then it's a lot easier before you build a bunch of things that will never get accepted.

So, and again, you don't have to publish it to that directory if you don't want to. But I think it's worth it.

It's not a lot of effort to get it done.

And the other thing is, is that, you know, you can go ahead.

My son's calling me and family comes first.

Okay, yeah, yeah. So, but yeah, I mean, yeah, the idea of pushing out there. And so, and if some of y'all are new to building software, once you start building something and sharing it, you kind of got to think about it like you've got customers in a way.

And think about them. Think about, you know, like my time in tech support and being on the front lines for years and hearing all the issues kind of gave me that empathy for people using the software that I was involved in.

And so you do need to kind of think about, "Hey, I've, like with Omastorm, I've built something. It's working a certain way.

And people are liking it." I would hate to then all of a sudden rip a feature out from under them or completely change the way the config works and then everybody's config breaks.

So you kind of got to keep those things in mind too as you move forward. But you don't have to, you know, go crazy.

And I would say start with something small that you personally have found use out of. Because an agent can write the code, but it is not going to know your intention for what you're building.

And it's not going to do as good a job of truly exercising it in the real way that it would be used. And so there's a lot of value in just building it, you know, running it, clicking it, going all over crazy with it, trying to break it.

Like that kind of stuff, it really adds a lot of value to a polished product. And thinking about, "Hey, I might be building this on personally, like an Omarchy Mac install, but somebody else might be on a, you know, an x86 base machine.

It should work in both." Because like with Omastorm, when I first built it, I actually was, I didn't have an Omarchy on a laptop.

I had it on a VM that was on an x86. And so the Rust binary that was built was just x86.

And I had people say, "Hey, I can't install this on, you know, the Arch one, what, you know, the Arm64, you know, but I did have my agent build it and it worked." And I'm like, "Oh, that's cool."

And then so took that information in and now I'm producing, you know, an x86 and the Arm64. So these are the kind of things you just got to kind of think about as you're building it out.

But yeah, are you back?

Yeah, I'm back. And one thing that, you know, as tokens get more expensive, and, you know, I don't mean that $200 is going up. I mean, you know, less and less.

Inflation, I suppose, applied somehow. One thing that I've been burned a couple of times and really ended up wasting tokens and maybe some time is I'll build, start building something and then, and I'll tell my agent, "Go look and see if somebody, you know, wait a minute.

Go look and see if somebody else has already done this better." And it comes back and says, "15 people have tried this and there's 17 that are better than you."

And so I'm like, "Damn it, I wasted tokens. I wasted time."

So, you know, wrap that back to the fork conversation. Fork it down, find the best one, fork it down and customize it to your own needs.

Yeah, yeah. In engineering and software, we'd call it like prior art.

Yeah, yeah.

You know, and so before agents and all, you know, you'd be coming up with a solution. And the first thing you do is you go figure out, has anybody done anything similar? You know, can I take the learnings that they, you know, had from it and apply that here?

So that's always a good first step for sure.

Yeah, and AI agents will gaslight you into making you think that your idea is the most unique idea out there. And then I've pushed back and been like, "Okay, hey, let's just see if anyone else has had this idea." And then like you said, Fred, it's like, "Oh, yep, there's 30 other applications that already do this and it's better than the process that I showed you."

So I think the thing I'm struggling with is I have an application I'm planning to launch on Windows, but like I keep messing around with it and finding bugs. And I don't know when the right time to actually, you know, launch it actually is to like have people, you know, like I'm, I figured my thought would be is when I'm using it every day then and not running into constant bugs, that's when I launch it.

But I don't know if I'll ever reach that point, you know? So I don't know if you guys can talk to about that.

Yeah, I mean, yeah, I mean, I would say number one, try to keep the feature set as small as possible to begin with. Don't be, you know, don't build, you know, all 15 features of what it is. And then you're, then it's like playing whack-a-mole.

You fix one, the other one breaks, right?

Yeah, now I'm way too deep in the features now.

I hear you. It's just like talking to a human employee. If you spend three minutes with them and tell them what you want, you're going to get three minutes of instruction results.

If you spend, you know, a day with that employee and really teach them what you want, you get better results. It's no different.

I treat Larry like he's literally, I mean, I start to kind of apply human attributes to him. And I've kind of encouraged him to, you know, push back on me and be snarky.

Larry's your AI assistant?

Yeah, Larry's. And there's a lot about how I built Larry at how.nixfred.com. And Wes, tell me if you're sick of the self.

You get one more and then.

I don't care if nobody goes to it, but if it'll help somebody, then good for them.

Yeah, no, it's all good, man. Cool. Yeah, so Zach, what's the name of your plugin?

Oh, so right, basically I have it on Windows right now. I don't, it's called Soto. So it's a voice transcription and it's not a harness.

It's like an agent management. So you can manage threads or have multiple terminals open to work with agents in.

And it's just a nicer UI.

So if anyone's familiar with T3 Code, it'd be similar to that. And I've built a lot of features similar to what T3 Code has, but I also have built-in transcription, which I've, that I've used a lot. So similar to Whisper Flow, but my plan is to port it over to Omarchy because right now it's an Electron shell, but I want to build it more native in a more native language to Omarchy so that it, I don't know, I see Omarchy developers as being preferring things that

are customizable. So I want the app to feel that way. So that's kind of what I'm planning to do next, but I'm just trying to get the Windows version up to my standard.

Then I'll release that and then start working on the Omarchy version.

That's awesome, bro.

Well, I think that kind of leads us in to open it up to some Q&A. So we're going to do, we got to keep it, we got a lot of people here. So we got to keep it to two minutes if we can per person.

If you got a question or you just want to talk a little bit about what you're building, that's fine. Just raise your hand and then I will make you or Fred will make you a speaker.

And then, yeah, we'll kind of try to get through the room. So who would like to go first or second?

Follow everybody, follow everybody. And I'll stick around after it's over to chat if anybody wants to chat, but Wes, you pick, man.

All right, looks like David is up. Let me approve David here.

Let's see, did that work?

David, I just accepted. Very good.

So what's up, David? What's going on?

Unmute.

Let's see.

Let's see if we can unmute him. I don't think I can.

All right, I might have to come back to you, David. Let's see.

Yeah, who's next?

Anybody else?

If you want to speak, raise your hand and.

Sorry about that. Can you hear me now?

Oh, there we go.

Awesome, David. Welcome.

I'm so sorry, guys. I apologize. So I've been following them.

I was using Omarchy probably right before Quattro, maybe about four or five months. Quattro kind of like just opened up the door with the harnesses coming and the skills to take a look at, you know, doing some things.

So written a couple of plugins, got them published. One of them's called HyperTile Equalizer.

The other one's called GitHub Repo Tracker. And there's a few of those out there.

So it's kind of like one of those things where you build something and then somebody else is already working on it. And you're like, "Oh, man, you know, I wish I'd known about that."

But I got the experience, which is cool. Question I have for you on the Omarchy plugin development stuff.

I mean, I've gone through it a couple of times. Works really well.

It takes a while to get through the approval process because the agents, you know, kind of go do their thing. And then it might take a little bit of time to get a notification that you got a security issue or something like that needs to be fixed.

I'm all cool with that. I mean, it's cool, but I really like to know like what is the shortcut to like pre-qualify or validate my own code before I submit it.

Wonderful question. And what I've done is I've had my agent go, because that, most of that information is like in the source code for the Omarchy plugin directory. And it's like, "Hey, go figure out the checks it's going to run ahead of time, right?"

And then look at what I have here and tell me where there's going to be issues. So yeah, I think there could be an opportunity here and people on here that are involved with the Omarchy team and stuff, y'all can correct me, but I think having an official skill or something that we could pull in and use, like before we even get started building the plugin or even after, it's like, how can we get to automatic, you know, almost automatic approval here or at least, you know,

cut that friction down? Because I do agree. It's kind of like, you know, if you haven't gone through it, it's like, man, I wish I would have known that ahead of time or whatever.

So there is some things you can do though, right, to have it go and pull that repo down and look at it. You say, "Hey, scan over the prior issues, you know, that have had security audits.

What kind of things did they fail on?" You know, are we going to hit those issues?

The one thing that is also just a little, I mean, like the first one was always painful, right? The second one was easier. On the second one, I think one of the tricks, I don't know if it worked or not, it felt like it did, was to say, "Okay, here's the plugin developers guide.

Here's the, you know, I grabbed anything I could from a link standpoint to just that was relevant. It was just a couple, but it was like, okay.

Relative to the context, let's make sure we're following the developer guideline and the submission guidelines and let AI just try to, you know, reassess it on its own. That seemed to work pretty well.

The one thing that is a little bit of a gotcha is like sometimes the agent will, like you've got to, you do your initial submission and it's like super sweet to just have it go do it. It's a little scary to be quite honest because it's like using a credential and stuff, but it does a good job of doing that first submission.

But then when there is a feedback that comes, you know, from hand course like, you know, automated, whatever, then you're like, "Okay, I got to go fix this." There have been times where it's like the solution is actually to go edit the anchor point in the submission versus just continuing on to say, "Okay, I fixed it and let it go forward."

It's like there's a specific sequence of things that has to occur for submissions versus verifications. Does that make sense?

Yeah, so one of the things that checks is the SHA, right? The unique identifier for the commit, because they want to make sure that the main commit that you have on your main branch matches the one that you had them verify. And so kind of the trigger to have it check again is to edit that issue and update that.

I haven't found any issue if I just tell it to edit the issue and update that. Like it seems to always pick it back up.

No, it's a good, it's one of those things that's kind of a learned thing. And it's also an experience thing.

I'm not as experienced with Git as you guys are. So it's like, "Okay, I'm trying to get my head around the plugin development, the languages, blah, blah, blah, what the AI is doing."

And then also what is the Git and GitHub specific stuff need too. So a good skill for that would be awesome.

If it doesn't exist, it might exist and I'm just not finding it or using it correctly. But it's been a blissful experience by comparison to other things I've done.

So I'm a happy camper and I'm going to let other people bring up their issues and stuff. So thank you very much for the inputs there.

Yeah, thanks, David. Appreciate it, man. All right, let's see.

Let me throw one more thing in on the submitting a plugin and little tips and tricks. So the first thing is don't submit it the first day that you build it. Even if you're happy with it, you're going to find something that you like to make it a little better.

So give yourself a week and work on it and sleep on it. I find that I make some of my biggest breakthroughs and ideas when I get the freak off the machine and walk the dog.

The second thing is that open your head up to allow PRs and let other people work on it, right?

So let other people, let other people join in the fun. And then the last thing I'd say is bring your patience because Handcore has got a lot on his plate and he, you know, he's got a lot on his plate.

And so those are the three things that I would say. Martin.

Yep, Martin.

We got a new speaker.

Yeah, hi guys. Hear me well?

Yes, sir.

Okay, cool, cool, cool. First of all, like awesome to be here and like awesome to see like what Omarchy has unlocked for so many people. This whole conversation that I've been following here like leads up basically to something that I wanted to share here in the group.

Like I built a couple of plugins, I think like.

And like when I was seeing the marketplace and the issue list that.

Is it bad?

Yep, we're good.

Okay, okay. So I saw the issue list, right? That Handcore is processing right now.

And I felt like, like, wait, this, this, we need to help him here. And you guys already have been talking about like the validation, like the stress that comes along with submitting your plugin and your work towards this whole process.

And this is exactly why I built like a tool called OmarKit, which basically does the whole marketplace pipeline locally on your machine.

So it tests those things that is doing. And it also comes like with some extra tools that I use personally, like because like I saw Handcore's list, but I also know like what AI is doing and like what the stuff is that I always review on.

So something that I wanted to do with OmarKit is like building the same test-driven development that I'm using to, or that I'm used to with. I'm a Larva guy.

So I wanted to have like something that can tell the AI, no, this is not well, this is not good. Yeah.

So also it comes like with a verify command. It comes with like an inspect command.

Right now I'm building a run command and that's actually, I think the naming is a bit wrong because like what I'm actually trying to do here, it's already released, but it's like a kind of an adapter that, first of all, it's like, it's handy for Handcore because Handcore always checks on like, what does your plugin do? Like, does it create a memory leak?

Does it do like all the freaky stuff, right? Like the run is like a kind of block, like an adapter that is attached to your plugin that makes sure that it cannot happen.

Those kind of things, OmarKit does. And like the subtitle before was like, it's a safe place to run your plugin and to test your plugin and to see if it's up for validation or, yeah, for the marketplace.

But right now it's actually, it's doing way more than that. So yeah, check it out if you have these kind of things, like a skill.

I saw like the skills coming up as Markdown files everywhere, but, and they are really, really nice, by the way. There's a lot of work being done there and I think that's really, really amazing.

Yeah, the only thing is like we all know, right, with AI, like skills go so far. Like I had so many countless, so many moments that I said, did you completely ignore what I just said here?

Yeah, I cannot count those ones. But so that's why OmarKit exists.

Check it out. Reach out to me.

I'm building it as we speak and it's really aimed to tackle these kind of problems you have.

Wonderful.

To make you feel more secure and more, yeah, yeah, that you can step up and like, let's bring this up.

I appreciate you building that and sharing. I think that'll be helpful. You know, something I was working on recently is like this idea of running it locally while you're testing it without having to install it fully.

And so, because I'm like, you know, I've done a lot of web dev and the idea of you have hot reloading and stuff where you can run it and see the web page reload while you're making the changes. And so having a way, like what I've done is just a Mees task where it's like Mees dev in it temporarily and like simlinks it to the plugins directory and then installs it and runs it.

And then when you kill the command, it removes it. So having something like that, I think would also be pretty helpful.

So maybe you could look into something like that or I could contribute to it, but.

Yeah, yeah. It's still not a P1, but it works. I saw already some guy from the community testing it out and it worked for him.

So that was good to see. But yeah, it's aimed to do exactly that and like to be like a hard guard for AI.

Like, okay, this is what I'm building right now. Obviously doesn't match the criteria that the marketplace is having.

Wonderful. All right. Well, hey, I'm going to give some time for other folks, but feel free to share in the chat and we'll catch you next time.

Thanks, man. All right, Colin, I think you're up next.

And hang tight, Thundy. Colin's next.

Thanks, Wes. Hey, I just wanted to, a couple of things real quick. There is a skill out there.

If you go to GitHub and you search Omarchy security skill, it comes up. And I'm not the developer of it, but I used it on two of my plugins with great success.

The user is JankEESVW. And he's actually got another great tool.

I think it was through him. Let me see if I can find it.

It's OV or OmaVM. And it's a tool that allows you to spin up Omarchy distros on Omarchy.

And you can use it to test your plugin.

Yeah. So those are two things that really helped me quite a lot. Of course, I've made a couple of plugins and I've only published a couple of them on the Omarchy plugins directory.

The others are on my GitHub. But, you know, I didn't want to specifically plug myself here.

I just wanted to tell you guys about the skill, which is very, very useful. And it got me through the pains of testing them out before I published them and saved me a lot of headache.

Because the first one I published, I just went through this constant, you know, the automated, you know, CI throwing back all the issues.

So, and I started, there was a conversation last week about possibly working on a local AI, doing a fine tune. And I actually started working on that.

And actually, I used that tool, the OmaVM tool to sort of test it out.

And if anybody's interested in working with me on that, it's currently in its state as a LoRA adapter. And it's kind of a difficult nut to crack for the, if you wanted it for widespread usage to have a local AI on your machine because, you know, you have to run on a potato.

So there's not many models out there that run well on a potato. So I did find one that was a very, you know, by LiquidAI that was apparently pretty good.

And for my basic testing, it's not bad. But I started out as just doing the training, the fine tuning, just to replace the default Omarchy skill.

And I want to let you guys know it's a lot more difficult than I thought it was going to be. You can certainly, you know, fine tune it and get it to do some stuff without the skill.

It's very easy to overfit.

So it would, the effort, if anybody's still interested in pursuing it, I don't know from last week, it would take a lot more input and training data to get it to be worthwhile. I don't know.

Yeah, yeah. I hear you. I mean, there's just so many man-hours that are going into those models.

It's kind of hard to compete with that.

Yeah, but I kind of set up an automated, you know, I had Grok sort of just, you know, doing automated runs and, you know, train, figure out what it needs to do to change, you know, to make it better. And, you know, you can pretty much automate the whole thing, but it's not, unless you have a lot of human input on the fine tuning training data, it's probably not going to be very good.

So, yeah. Well, hey, thanks for chiming in, Colin. I'm going to move on to Thundy next.

Evening, gentlemen. Can you hear me?

Yes, we hear you great, Thundy. I got to know, man, how did the London meetup go? You're the leader.

I want to hear more about it. We're trying to build one plan and you had great success.

So what's the secret, man?

Oh, the secret, I wish I could tell you, but we've not held it yet. It's happening on Thursday.

Oh, look at me.

Well, then prophetically, what is the secret?

Well, you've already got more people there than prompt committed. So what's the secret?

Secret, honestly, I just did it. I didn't overthink it. I saw the post from DHH where he exchanged a WhatsApp message with, who was it, Toby, when they were raising funds for OmaCon.

Immediately I saw his post. I was driving, I just parked the car and I called my friend. I said, there's this thing.

He hadn't even heard of it. He said, Oma what?

And I just said, you just trust me. You give me a venue, I'll sort everything out.

And they happily gave us the venue.

Wow. You just did it. You didn't think about it.

That's the motto, right?

Yeah, which is very unlike me.

Hey, that's awesome.

Changing people's lives.

That's right.

I had a quick question for you, Wes. This relates to a question you asked earlier today about the Rust repo.

I just wanted to ask if you could explain to us, because when I saw it, what came to mind was maybe Wes wants examples so you could steer your agent if you're building anything or implementing anything in Rust to say, follow these principles. I assume, but I didn't know. I just wanted to ask you if that was the case.

Yeah, because, you know, I've been doing software 25 plus years professionally, you know, and so I know a lot, but Rust was one of those things I never touched. It's newer. And so my whole thing was, you know, the way I approach things when I learn new languages is I go to find those exemplary code bases that not just look pretty, but that are performant, that have worked well.

In this case, open source, and it seems to be a healthy, you know, repo. And so the idea was, yeah, I want to point my agents at it to say, hey, go look at this code base, pull out the patterns that are being used in the structure.

Where do you find, you know, things that work well and things that have been pain issues? And then the idea from that is kind of to distill it down into a style with the style being examples of code.

I find that when you write a lot of rules in these agent docs to say, you know, do this, don't do that, it doesn't work as well as literally just having code examples. So like in the rails world, show me an example of a well-structured controller model, right, and views.

Don't tell me the right things to do in those. Just show me good examples.

And it tends to follow that better.

Understood. Thank you very much. That's everything.

I don't want to hoard anybody's time. Thank you so much.

Yeah, thank you. Good to hear everything is moving forward and good luck this week. Anybody else want to speak?

We've got time for probably two more people and then we'll call.

Is Heath out there?

Oh, it looks like Hancour has hopped in. You want to give him?

Yeah, let's talk to him first.

You got anything to say or you just want to hang out?

Request to speak, Han. There it is. I got it.

Request to speak, Hancour. And Heath, hang on for just one second.

Hancour, if you want to speak, go ahead and request to speak and we'll move on to Heath. Everybody likes super sexy hardware and we want the biggest GPUs.

Heath has been a buddy of mine for 20 years and he, I can't say where he works or what he does, not trying to be cool, that's legit, but he gets to touch the big stuff. Heath, welcome to the show, man.

Tell us about what you work on.

Hey guys, how are you?

Doing fantastic, man.

Yeah, so this week has been an interesting Fred. You and I were chatting last week, you know, with Wes and everybody was, you know, I'm fairly new to Omarchy and so I'm learning the plugin aspect. So I got a lot of projects that, you know, I'm trying to like figure out how to do.

But one of the things I kind of indicated last week and I made a little bit of progress, but because I'm a former AMDer and I got the new ROCm10.0 actually working on my Omarchy Linux AMD system here. So I had to cheat to get it to working first just by spoofing the ID.

But now I'm going to work on actually creating an Arch Linux derivative underneath so that it, at least from an Omarchy perspective, to try and make sure that it runs cleanly. Because the new ROCm10.0, massive performance improvements in the AMD world.

And then, Fred, I was tweeting out to you is that, you know, I recently just picked up a, it's the Dell equivalent of an NVIDIA Spark. And so it's called the GP10.

And so I'm just playing around with it with the default DGX OS right now. But I found somebody that's already done some of the work to get Omarchy running, you know, on a Grace Blackwell chip because that's an ARM CPU matched with NVIDIA GPUs.

And so now I'm just going to do a little bit of a benchmark to see where Ubuntu, DGX is a Ubuntu derivative. So I'm going to do some benchmarks on that.

And then last but not least, my long-term goal is that because I'm so tied into the federal market space, I'm going to try and create some shortcuts for any federal customers that would want Omarchy in the future. So that means I got to go out and create a crypto validation suite that can go run through the past, you know, go get tested at NIST.

And that's going to take probably the better part of a year to do that. But in the meantime, the way that we would do it in the federal space is that we would get the updated list.

And Jesus, Fred, I'm drawing a blank on what my own stuff is for the checklist to be able to go out and just, you make these settings and then it's considered safe for running, somewhat safe for running inside of a federal environment.

Omarchy Fort Knox. We'll call it that.

No, there we go. Yeah, we can use the Fort Knox. And so I have a lot of projects in front of me.

The problem is that it is the end of the federal season and I am like crazy nuts busy. So I don't know how much work I'm going to get done over the next three weeks maybe, but maybe see if I can take a week off once I get there.

What's the biggest box you've put together that you've gotten to play with, like the most powerful? We all like big and strong, but most of us don't have access to that stuff.

I've been on about a 5,000 GPU system. And then I've also been on a, from a supercomputer perspective, I've been on a probably about a 250,000 core CPU system, something like that.

There you go.

Yeah, I mean, that's my job. I design high-performance computers and AI systems for the federal government. So I get to talk to all the Department of Energy people all the time and, you know, kind of do some big projects.

So, yep.

Kind of like Tony Stark a little bit?

No, not me. Not me, man.

He's very humble, but he's super smart and so much fun to be around. Thank you for joining, Heath.

What's that?

I said, you're so much fun to be around and a great friend. Thanks for joining today. And Wes, I'll hand it back over to you for our closer.

Yeah. Yeah, thanks, Heath. Yeah, Hancour, nice to meet you virtually.

We're just, this is our second week of doing this. We're just chatting about the whole plugin ecosystem and, you know, some people have backgrounds in programming engineering, others don't.

And so it's just been a really cool environment to be able to get our ideas out there, to have something, you know, that we use locally first and then decide, hey, we want to share this with other people. And so we've been kind of talking through what that process looks like.

So yeah, first of all, thanks for building the marketplace and all. That's awesome.

But yeah, anything you want to share, any new updates coming? Are you swamped in issues and PRs or any way we can help?

Well, first of all, thank you for the invite. I just joined because of Nix told me, asked me to join.

Thank you for joining. Hey, Hancour, I don't know if you know, but there's this place that you can register and submit a plugin.

Hancour runs it, everybody.

Yeah, I did plenty of updates today. They were necessary, to be honest. And I didn't have a lot of time to take care of the marketplace.

So maybe some people recognize that I didn't really do a lot of improvements the last week. So yeah, I just hopped today in and, yeah, started to do some maintenance and yeah, that's it, basically.

Wonderful. Yeah, because we were kind of talking about the process of like, you build a plugin, you submit it, you create the issue, and then you get back security, you know, audit validations and how to work through that process. And we were kind of talking like if we could know ahead of time what things would be checked, you know, you could have your agents kind of keep that in mind before you even build it.

And so we had a guy share about something called OmaKit that he built that kind of helps with that. Somebody has built a skill, but is there anything in the official marketplace or anything in the plans to kind of like beyond the, you know, publishing the plugin guidelines that we could use?

Yeah, and what's your process? How do you approve them?

Sorry again. I didn't hear you.

Go ahead, Wes.

No, yeah, no, I think Fred was just saying, what's your process for approving? What's that look like?

Yeah, well, those were two or three questions. Well, we have some plans like for the new marketplaces, which eventually Ryan or DHH will announce them. So I will not talk much about it, but there will be new submission system as well.

And yeah, it will become soon, but at the moment we are building it or Ryan is mainly building the backend. I just supported him in the frontend for the webpage.

And now it's taking some directions, I would say.

Wonderful.

I don't want to spoil a lot of it, but that's the current state. And how to submit, well, there was some like the OmaKit. This is one tool that you can use.

And one guy also shared one GitHub. I can't remember the name exactly, but this skill also helped a lot of people to get validated and verified for the plugin.

So yeah, I would recommend to use this. I cannot predict how the new market system will be checked.

There will be, of course, checked by the agents, but how deep, how far, what's the requirement for this, I don't know yet.

Well, hey, I'm just excited that there's, you know, there's some movement happening and that people are working on it. So that's great.

Wonderful. Well, thanks for joining and I don't, let's see, anything else you want to say before Fred and I close out?

No, just thank you again for the invite.

Yeah, thank you for joining. Appreciate it, man. All right, let's see here.

All right, so good chat today. Looks like we got one more person in line and then we're going to call it a day and we'll get started with our week.

Okay, let's see here. It just approved.

Mihai, is that how you say it?

Mihai

Yes, can you hear me?

Yes, I can. What's up? What's your question or do you want to share something you're working on?

Mihai

Yeah, I wanted to share what I've been working on regarding Omarchy. So I managed to reverse and get an Unreal Tournament engine working on Omarchy.

Mihai

So I got a game fully running that's running on Linux as well, but basically on Omarchy, on the Mac. And I got it also on the MacBook, which was quite a bit different.

Mihai

So there were quite a lot of differences between them. It's quite interesting.

Mihai

It's fully running. And I find it quite interesting that it runs pretty well in Omarchy with the current kernel.

Wonderful. And what's your favorite game to play in it?

Did we lose you?

I think we might have lost him.

Maybe lost him. Okay.

You want to try Tech Luddite and then close it out or what do you think, Wes?

Yeah, two minutes Tech Luddite and then we'll have to close it out.

And then we'll have to see how bad if my buck's lost or not. I don't know.

It was close when I joined.

Well, it doesn't look like he's raising his hand.

Okay. All right, well, let's just call it then. I appreciate everybody joining.

It's been another fun time, Sunday social with my buddy Fred here. And yeah, we'll keep doing this.

We'll keep having conversations. We might have more special guests.

We had special guests today. Maybe we'll have more later.

Everybody's a special guest, to be honest with you.

I love that. That's exactly right.

You know, because we're all, you know, we're all in our own places and our own life building and just moving forward and having fun. And let's let this be a positive place to share and help each other.

And follow each other.

Yep, let's do that. So let's call it. You all have a good week and we'll see you next Sunday.

See you next Sunday. Thanks, Wes. Thanks, everybody.

Yeah, thank you, Fred.

All right, I'm clear.

↑ back to the summary