Coding is Solved. Next Up: Figure Out How to Plan

Let me tell you about it. So show of hands because I love engagement from everyone. How many of us have 2 or more agents that are, like, accomplishing real useful things Sn an always on way? This could be like a claw or something, and it can shut off. Keep keep your hands up. Keep your hands up. Okay. Great. That's a considerable number of us. How many of us have our agents talk to each other? Okay. How do they do that? Go ahead and shout it out. Throw by. Say it again? I'll talk you'll see my demo. Okay. Yeah. Okay. So you do that. You do okay. Notify. Notify. Right? Like, your message sending between each other. Maybe they're writing markdown down files, right? How many of us talk to our agents? Okay. How do you do that? Because you may have a huge chain of stuff that's going on. How do you insert yourself into that chain Sn a critical One this critical time? Stop the agent, write the prompt, and then repeat it. Message. Yes. Yes. It's chat, right, usually, if we're lucky enough to catch the agent. Sn, you know, you start thinking like, right now all of our agents are on our local machine or maybe, you know, they're in the Klaus, but like really they're on our local machine and they're writing a bunch of markdown files. One end up with this like graveyard of markdown files and it's, like, half baked plans and and stuff that your agents Managing of share. But, like, what if you really care about jumping into a plan point by Platform, like, reviewing an output point by point? You're stuck in that same, like, chat window where you can talk about 1 thing at a time. And I find myself, like, copy pasting from the markdown file, like, hey. This thing, can we make it more like that? Or that thing, can we make it more like this? And there's gotta be a better way. Sn, you know, One then I kinda combine that with I'm a software engineer during the day, and I use a lot of agents to write a lot of code. I'm sure a lot of us can empathize. Like, a lot of what my job is becoming is something like an AI sandwich. And that means spending a lot of time in the planning. I'm seeing a lot of heads nodding. That's good. And a lot of time in, like, the output because the the harness and AI and and MLM is so good at like executing. It has a really good plan and I just wanna see like what came out on the other side. So like what's really good at that? Documents are. So I wanted human One humans and agents to be able to collaborate on this like shared thing in a way that felt really native to both, in a way that like current text editors aren't. So maybe you can get okay. Thanks. So so maybe you can get, you know, write a document all at once or something like that. But, you know, you can't do the kind of primitives that humans can do. So I think agent Data docs are kind of a new primitive that we One to like think about how it fits into our workflow One like what gets unlocked for us. I have a lot of ideas. I have a lot more like kind of coming that I can't like I'm not ready to talk about yet, but I but I wanna hear from you. So if you are interested in this at all, I would love for you to go to comment.io. There's a little Discord link. I'm not on socials. I don't know how to do that. But like, just come talk to me in Discord, please, because I really wanna understand how you can imagine using this. Now the the code part. So, by the way, I'm demoing this in comments.io, in case you couldn't tell me. This is, all on CloudFlare, Durable Objects. My understanding is that it's fairly infinitely scalable. I've been doing, like, back end scale work for a long time. As far as I can tell, if it's like 1000000 dollars would cost me, like or no. Sorry. 1000000, documents would cost me, like, $10.15 dollars a month to host. So, like, it's kind of free forever, this thing that I'm building. And, you know, I'll put all these kind of architecture documents in the Discord for anyone who wants to take a look at it. And and and by the way, this is all editable by agents. Right? This is just a mermaid diagram that my agent wrote. By the way, just showing, like, 1 cool feature is so, like, this text, like, 1 thing that comes up when you have a lot of agents working together is, like, who wrote that? Like, what's going on? And so I built a thing that shows you exactly who wrote what, and you can have a lot of different agents, work on this. And I built a thing that is still very early Staff, but where you can tag in your agents. And this is going to my claw. There's a way to do it through codecs or or quad code as well. And this is gonna take a long time because it's running on Opus, so it's gonna and it may not be till the end. And then, just a couple of things I wanted to talk about, like what makes agent native work really well on the web. I'll be really fast. Okay? 1 thing is if you want to know if an agent is making a request on the web, and not a browser, you look at the accepts header, and it's not like the, Agent. And and and the accepts header will be something other than just HTML. Browsers only accept HTML. So what that allows us to do is when you copy paste One URL from comment.io to an agent, it actually sends down a full, like, JSON representation of the document and a, like, docs for, like, how do I interact with this. And and, I'll stop there. It's 1 of your things, but I'll put it into this work. I'm happy to answer some questions. As of what I'm answering, help me break my editor. Please go to this Real link and collaboratively edit One I wanna get all the errors in the world. It sucks right now, but I'm gonna make it better. Just for a few questions, do we have Sean Ransom in the room? Yeah. Oh, Yeah. Sn it all comes down to how Google Docs does it is like conflict free replicated data types. And so that's like individual people are like making these little messages that then kind of all get mushed together. And so that was I'm glad you asked that because that was literally my second thing is like you give the agents the tools that they know how to use. So how do your agents edit their markdown files is with an old string and a new string, like replace this string with that string. And so that's the primitive that comes up in the API. You turn that into a CRDT Sn it can seamlessly merge with with everything that the humans are doing, and then it all just kind of flows together. And again, you know, there's comments. There's a cool thing, like, you can lock documents and only get input. This is I can't believe this hasn't broken. That is wild. This is, like, so new y'all. My product is working? No. I mean, it's, like, you don't understand. I was changing so much this afternoon. So, yeah. Please come to the Discord. I wanna talk to you. Thank you.