Vibe Check: Claude Cowork Is Claude Code for the Rest of Us

[00:00] If you're a non-technical person, [00:02] You are used to a world where [00:04] You send a prompt and then you get a response within a couple of minutes. And once you send a prompt or a chat, you can't do anything else with that AI. This is built for working with your AIs in an async way. This new Cloud Cowork app is a really good example of Agent AID of Architectures, which means at the bottom of the app, instead of having like software that works by deterministic rules, you have an agent and the agent is wired up to the UI of the app. I've been at the topic for a little bit, but this is the product that my team has built here. [00:34] at this for the last week and a half. [00:36] And [00:37] What we're trying to- The last week and a half? [00:40] That's it? Come on. [00:42] We've got a new... [00:44] Anthropic drop. [00:46] So, [00:47] Anthropic just dropped Claude Cowork. [00:50] which is [00:52] basically cloud code for non-technical people. We got access to it early at every. And so I'm going to give you a quick run through of [01:00] what it is and how it works. We will have a full write up on every in a few hours. Probably we're just we're kind of figuring out [01:09] how to do these things. So [01:12] And I'm going to add Kieran. Kieran is here. Hello, Kieran. How you doing? Hey, what's up? [01:17] Thank you. [01:18] So I'm just telling everybody, if you just got here, we are about to do a vibe check of Anthropics new Claude co-work feature, which is a basic cloud code, but for non-technical folks. I'm going to just demo it for you right now in here.

[01:35] And [01:37] Okay, so let me just show my screen. [01:39] All right. So this is what it looks like. It's Cloud Cowork. So you'll see [01:43] We've got the chat over here to the left. You've still got regular chat. You've got code here, and then we've got co-work. So it's three Cs. [01:51] It starts with, let's knock something off your list. I love the copywriting here. You can tell from the way they did this that it's really designed to do deeper tasks on your computer than maybe chat is. So create a file or crunch data or make a prototype, send a message, organize files. [02:15] It's got over here, progress. It's got artifacts, context. We've been playing around with this for a couple hours now. I think it's cool. I think it's cool. I think there's a lot of really interesting questions from the UX perspective of, [02:29] chat versus code versus co-work. [02:31] And I think you can really think of whole work as being, it's like chat that has access to your computer and runs for a long time. [02:39] which is essentially flood code, but just less intimidating. So a couple of things that we did that I think are interesting to look at, [02:47] Okay, so here's an example of it working. I asked it to go to the every.to website and find five competitive companies that do the kind of consulting that we do and then analyze our positioning. And you'll see it like just went and used my computer. It's running in a long loop. So this is a really interesting one where...

[03:09] You can do this with regular Claude, like Claude can do this, but the number of iterations that it's going through, this is many, many, many minutes of iterations. [03:17] Thank you. [03:18] so it looks a lot more like um actually can you make this mark down so it looks a lot more like the cloud code but it's friendly enough for anyone to anyone to use um well we'll look at the [03:31] Another thing I had to do, I have a dinner that I have to go to tomorrow night that I had to prepare some remarks for. And I asked it to basically like go to my Gmail and prepare remarks. It has a connector to Gmail and to Google Calendar, but the connector wasn't working, I think, because this is like very beta and was not out when I was testing this. [03:55] And... [03:56] Let's see. And I asked it to draft a response. And it drafts a response. And I think the response is actually pretty good. [04:02] So... [04:04] And this actually sounds like me. This is kind of crazy here. And you've got to look at this because this is-- that's some implications for Korra. [04:11] Thank you. [04:12] So basically the setup was I have a dinner tomorrow that I have to do remarks for. [04:16] And [04:17] I asked and the whole thing organized at dinner was asking me, um, [04:22] Can you tell me what you want to talk about? [04:25] And so I just said, like, find the email and draft a response based on what you think I would say. [04:33] And, [04:34] This is something that I think I actually could send with minimal edits, which is pretty cool. Are you seeing this, Kieran?

[04:40] Yes, looks good. I think one of the things that's good about this is it's gone through many, many steps to both identify what I would say and how I would say it and has all the context, which is like, yeah, it's pretty cool. [04:54] Yeah, go for it. Yeah, so how is GoWork different than Chad? Like GoWork is more made to [05:01] You can share your screen, but it's more made to [05:05] like go longer, like really work on something. So it's more focused on getting you some work done. And this is what we know in Clothcode already, like when you trigger UltraThink or your trigger planning modes, things like that. So [05:20] It is similar in certain ways, but it also will unlock a lot of things that you use as a developer in CloudCode that now suddenly work very well for non-developer tasks. [05:31] Yeah, I can sort of see this being a thing that just really accelerates, you know, our growth team or, you know, our people who are in our consulting business who want to do some of these tasks and maybe are using cloud code, but maybe it's like a little bit less. [05:50] Intuitive. It should be released now. I think that they're maybe holding it back, but the blog post is live so you can take a look at the blog post. It's a co-work research preview. I'll throw it in the chat. And [06:06] Yeah, so I think it'll be it'll be out soon. Let me I'll like I can check with my anthropic people. But okay, let me go back to this. So I also asked it to do a calendar audit. It looks like it's actually still running on this, which is crazy. I asked this like an hour ago, go through the past month of my calendar and do an audit. Tell me how this reflects my priorities and whether it's aligned with my goals.

[06:29] Can you share your screen again, please? Oh, shoot. Yeah. Thanks. [06:35] Yep. [06:36] uh tell me how go through the past month my calendar and do an audit tell me how this relates to my goals [06:43] And it just, you know, it's been browsing on my computer for like, [06:48] hours and hours. One thing that's not hours now, but like for about an hour. One thing that's different about this, which I think is really interesting is [06:55] In Claude, when you start a chat and it's responding, you have to stop it in order to send a new message. But this, I can just add to the queue. So this is a little bit more like the... [07:10] This is more like the cloud code experience. So I think this is this is one of those situations where [07:18] it's built for you to send stuff as it's working. And it's not like a [07:22] like one message, one response, one message, one response. So it's really more built for long tasks. [07:28] Yeah, and it has the to-do task also built in on the right, which is nice. So you can see where it is and what it's doing. [07:37] Yeah, exactly. Another thing I did is I fed it a book and I this is a book that I it's called the outsider that I've been reading for a book that I'm writing. And I just asked it to like basically read the book, read the entire book and construct a taxonomy of all the main characters and ideas. And looks like it did this. And this is something that you could also do with with the.

[08:03] with regular Claude, but it would just be less detailed. This is really interesting. I wonder. So yeah. So if you want to trigger those longer running tasks, like, [08:16] you can just say, make a plan, do this. So I assume if you just push it to do that more, it will just run for longer. So if you say, take an hour, read through every single email, it won't give up as easily as before, and it will just keep going. So please... [08:32] Try those things. Yeah, it has a... It does have a plan mode. It said I have not used the plan mode yet, but that's pretty cool. I also asked it to go through our post-hog analytics and do some data gathering. So we published this guide. If you haven't seen this guide, it's like I keep talking about this. Everyone and everybody is making fun of me because I can't stop talking about agent native architectures. We published this guide last week. And we have these... [08:58] we have these uh buttons read with claude reo chachi bt and i was curious [09:03] Okay. [09:04] how many people click these buttons. And that's the kind of thing where I would go ask Andre, who runs our platform, and be like, can you go look this up? [09:12] But instead I just said, can you go into post? I just went to co-work and I said, can you go into post? [09:17] post hog and just find all this stuff and it said okay total chat with claude button cooks 4 000 that's actually freaking crazy oh my god we should get money we should referral referral fees baby uh what about uh chat with chat gbt what about um

[09:37] copy for agent. So this is cool because we were in a rush since we started this morning and we didn't have any MCP set up. What we just did was we connected Chrome and Dan is logged in on Chrome in PostHawk. So just browse through the thing, got the things. So that's very handy. Like MVP, just make sure you connect Chrome. That's a very good one to add already. You can also do that with the normal, but it's very good at browsing [10:07] figuring things out, especially [10:09] now it doesn't stop as quickly, which is really handy. [10:13] So anything you can do in a browser, you can now use co-work for to like [10:19] use longer running tasks and kick off things and it can [10:22] Use multiple tabs as well. [10:25] So you can have five tabs being controlled by Cloud and running, which is great. [10:31] Thank you. [10:32] totally totally and it sounds like uh it's this is down for people so come come hang out for me yeah it's down for me as well but but for dan is working so i'm the only one it's open for so we've got a monopoly here on on anthropic co-work content so yeah you're here and you have stuff you want me to try uh just let me know i'm happy to happy to throw it in the chat and um and make sure [10:59] All right, I have something fun to do. Make sure that you read every because every is the only subscription that you need to say at the Edge of AI. Every.to, we've got these vibe checks when new models come out. We get them beforehand. We had this several hours ago. We knew it was coming since last week. And we always have all the up-to-the-minute stuff that you might need. We also have

[11:23] A bundle of apps that we make. We have an app like the one that Kieran makes called Quora, which is an assistant, an email assistant. We've got one called Sparkle that helps you organize your files. We've got Spiral, which helps you write. And we've also got Monologue, which is a speech to text app, which I will show you shortly. And let's actually go back to... [11:43] Let's actually go back to the Claude demo real quick. So... And Kieran, if you see anyone asking questions, just... Yeah, there's one good question from Hunter. He says, "How good is it at research?" Like, one of the things I love Claude's normal for is the deep research. Like, does that exist? So maybe we can see if it can do deep research on something. [12:09] Yeah, let me know if you have a research query you want me to try, but I can show you... [12:15] you know, [12:17] Actually, you know what, I'm going to share a different screen. Hold on. [12:21] Um, plus... [12:23] Share screen. Okay. I'm figuring out my live stream setup. This is the first time that we've really done a, uh, [12:31] live stream for a vibe check okay so um [12:36] Now you should be able to see my screen again. So for research, like, okay, it depends what you mean by research, right? Because this is a research query. Can you connect to post hoc and tell me for the agent native guide that we published last week, how many people clicked the chat like that is research and it does give me [12:49] actually like a really good, um, [12:53] Answer.

[12:54] Which this is so interesting. Okay. Um, but like another form of research that we talked about is okay, analyze my competitors. [13:05] This is specifically analyze our competitors for the every consulting business. And I'm gonna say open and proof. Proof is the agent native Markdown editor that I built over the weekend. Can you believe I just said that? And so this is a, [13:20] This is a research document that [13:23] I think it's a good thing. [13:24] uh claude co-work put together [13:27] um which you know it's not like it's not the meatiest thing i've ever seen [13:32] Let's see. [13:33] I think this is not bad. [13:36] I'm not noticing like a super significant difference between this and like what a normal quad would would pull out. But I can see the research itself is more. Um, [13:48] is much more extensive than normal Claude would do. And [13:53] Let me just actually throw this into normal. You can see also, so if you do deep research, the research agents in chat, that's pretty extensive normally, but here you can see more. Yeah. [14:08] So, [14:09] Yeah, I don't know. Maybe that is also available in that version. We're still figuring out what everything is that is available, but it's very close to what you can do in Cloud Code. So the kind of fusion of chats and Cloud Code is co-work.

[14:25] Yeah. Yeah, this is very like you can see that they're calling these tasks as opposed to chats. So it's supposed to be, I think here's a good way to think about it. [14:34] A good way to think about it is [14:37] If you're a non-technical person, you are used to a world where you send a prompt and then you get a response within a couple of minutes. And once you send a prompt or a chat, you can't do anything else with that AI. You have to move on to something else. This is built for working with your AIs in an async way. So everything is set up, like the idea of a task, the idea of having a queue. This is all set up so that you can say, go do something and then not think about it for a while and then come back, [15:07] from Claude where the normal Claude app, you're kind of, you're trying to [15:11] You're trying to get an answer pretty quick. And I think that's the best mental shift. I think the real question that I have is... [15:19] is this deserving of its own tab? One reason it might be is there's a difference between like how you might treat one of these versus one of these. Like these are more throwaway. These are probably like bigger chunks of work. [15:33] But honestly, it's kind of confusing. I would rather just... [15:38] I think that they, I would rather just have it all in one tab and then have it, you know, [15:43] do different levels of research and thinking and and async based on the task and maybe based on [15:51] you know, a setting, like saying, like, really fucking think about this. I don't know. What do you think, Jiren? How would you solve this? Yeah, like, for me, it's also confusing. It's like, oh, great, there's another tab, and I have to, like, first think where to go. But I do get it, because I've seen this transition as engineers, as an engineer, like, we had the, like, copy-paste into ChatGPT, obviously, and then, like, that evolved into cursor, like, more agentic.

[16:20] which evolved to like, I don't look at code anymore. And I think [16:25] there will be a similar transition for people that do research or co-work. Like maybe now people are used to go into Chrome and like seeing what's going on, what's happening. [16:37] towards more of like [16:40] I'm just going to let it rip and do its thing. [16:43] And then [16:44] after it gets back, I'm going to review whatever the output is rather than understanding every single step all the way. And then you're like clearly already there. That's how you're thinking. But I do understand if you're not there, if you're more like in the chat thing where it's like, oh, do this now. Why don't you look at this? Like if you have this conversation, [17:03] It's maybe more chats and co-work is more you hand off a task to your agent. And your agent comes back and you review the work and you can follow up, but you can give extra directions. But I do understand to introduce a new tab because you need to shift your mind to do that. And obviously we're... [17:21] doing that. But I understand lots of people in the world, especially that are not coders, still have to make that transition where you just hand off something and then let it do stuff for 30 minutes or an hour and then [17:35] come back and review that. So I do understand why they want to separate it, even though it's all the same technology, because chat, code, and co-work [17:43] it's all the same model and it's very similar harnessing around it.

[17:51] It is philosophically or how you use it maybe a little bit different. So I guess that's why they did that. [17:57] Yeah, it's interesting, too, because when we did this... [18:01] when we did this [18:02] original vibe check when we got it a couple hours ago we had it we had me and kieran and a couple other people on the phone from internally at every and we were demoing it together and their initial reaction was like i don't know how this is different and is this like even that [18:14] that useful compared to regular Claude or Claude code because a lot of them are just using Claude code directly and [18:21] I think that that's like that was a really interesting thing for me where [18:26] you're not going to actually realize how useful this is until you get your hands on it. And there's probably going to be a learning curve. [18:31] on it where if you're a non-technical user who is used to [18:37] Who's not used to the idea that you can just like hand off your work and then come back. It's probably gonna take a while to like actually figure that out and get used to this as a UX paradigm. So maybe there's some... [18:47] benefit then to like having it be a separate tab so people can like basically realize, oh yeah, this is, this is different and I should treat this differently. [18:55] It's a real adjustment. [18:56] Yeah. Yeah. Yeah. And that's, so if you want to learn about that adjustment, like we're writing about that for coding, but you could really apply. [19:07] whatever is happening to coding, probably for co-work as well, like how that shift happens, how that goes. So yeah, like, [19:18] We thought about, we wrote about some of these things, and I created a plugin around this idea. So what I will do in the coming days as well is like see if the...

[19:28] pattern or the paradigm of compound engineering if that applies to co-work. Can I get that working in co-work? Because I would be very curious to expand that and see if it works inside here as well. Anthony, I see that you're from Anthropic. Do you want to come on the stream? I'm going to send you a link. [19:49] I would love to hear what you have to say about this and anything that we should try or anything that's missing. [19:57] Here, give me a sec. Copy. Let's see. All right. I sent you a stream link. If you want to come on, feel free. If you're not feeling it, that's also totally fine. [20:07] Yeah, he says it looks or is closer to Claude's codes than chats. And it feels like that because all the tools like the ask user question tool and stuff like that have a UI, which is nice. So you can ask it to say, hey, can you interview me, ask me a few questions, and there's like a nice UI with multiple choice and stuff like that. So [20:29] Yeah, it's really cool that you wipe on it. [20:32] It is really cool. [20:33] Um... [20:35] Okay, so let's keep looking at this. [20:40] It's still working. One of the things I noticed is when it was erroring, it had a gave an error from cloud code, like the internal error messages cloud code. So it's it seems like it really is just like a UI wrapper on cloud code, rather than a different agent harness, or maybe like a cloud SDK. Anthony, if if we're wrong about that, or anybody else from Anthropic, who's who's listening, I'd be very curious for your for your thumbs up or thumbs down on that. But that's like interesting

[21:10] be in the app. So it's like pretty easy. You don't need to use the SDK, but it's a really interesting thing to see. [21:15] One thing to... [21:18] One thing that I want to show people in case you have not been [21:22] watching everything that we're doing at every and if you haven't been I don't know why you would not I don't know I don't know what's wrong with you but one thing that's really cool that we're that we're thinking a lot about is agent native architectures. [21:33] um and an agent native architecture this is this app is a really good example this this new this new cloud co-work app is a really good example of agent native architectures um where we think about agent native architectures as sort of like cloud code in a trench coat which means at the bottom of the app instead of having software you have like software that works by deterministic rules you have an agent and the agent is wired up to the ui of the app and so when you click a button [22:03] with a prompt. And I think this is a new way of building applications that we've been working on internally at Every. [22:11] if you're interested in that I highly recommend that you look at this guide. We'll put a link in the chat. But this guide goes through how to use or how to build agent native architectures. It's like, it's pretty cool. And it's like it makes you build stuff like I built this, this is a markdown editor that I built with. This is a markdown editor that I built with cloud code over the weekend. So in the last couple days, I built this whole thing. It helps me track we use

[22:41] uh [22:43] when I get a plan from Claude, it helps me track, okay, what things have I approved? So you can see I'm approving things here. So I can track what have I approved? What have I looked at? What's what's done? What's not done? And yeah, there's a lot of there's a lot of cool stuff here. I'm gonna stop blabbing about agent native and go back to Claude. Karen, anything you want to add here? [23:01] Um... [23:03] Really what I'm curious for is like as a power user, can I load my own plugins and stuff like that? And there were questions about can we do custom MCPs? I think all the MCPs you can do and also this app has access to your machine so it can use AppleScript to load things on your machine which is really cool. [23:24] Yeah, and I'm really curious for where does it go? Like, clearly it's trying to get cold codes to the [23:33] normal user. But I think as a power user, I would use this as well because in the morning, I start up Cloud Code to make a daily planning. Like, well, what am I going to work on for the day? But it feels... [23:45] It's probably nicer to do in an app and also if this translates to mobile, this is super powerful because you can do more powerful work on your phone that's not necessarily... [23:56] code related. So [23:58] Yeah, we need to experiment, but there are lots of interesting [24:03] I'm very curious as a power user, even though it is the same technology, maybe it's a new way to use it, which is interesting. [24:11] Yeah. So this is my calendar audit that I asked it to do. So I asked it to audit my calendar and compare it to my goals. I reviewed your entire month of calendar activity. So this is something that like regular Claude probably would not do. I have a lot of meetings and standups. So that's an interesting one. I actually don't go to a lot of these. So that's probably not fair. And I also have a lot of one on ones. I have a lot of podcasts scheduled, which I'm starting to get rid of. So it

[24:39] Content media, health non-negotiable, blah, blah, blah. [24:42] Uh... [24:43] Many days have 10 to 15 plus scheduled events. That sounds [24:47] Probably right. [24:50] This is interesting. It said, what are your top priorities? I would expect Claude to know this. I wonder if it has access to all my memories yet, because Claude definitely knows what my priorities are. But anyway, this is pretty cool. I like this. [25:04] um see if it did oh we got we did we got the taxonomy it's it feels like it's it's like slowly lazy loading all the conversation history so it doesn't [25:15] it's like, this isn't happening. This is already are this already done. It just didn't load it. [25:19] I feel like a lot of the affordances here [25:23] They haven't built the statuses yet, so it's easy to see like, you know, and [25:27] In the code tab, for example, I could pretty easily see [25:31] usually what I've merged and what's waiting for me and stuff. But this is like just [25:35] an unorganized list, I guess, by recency. [25:39] or it's not there's no visual differentiation which is interesting hello Kate [25:43] Our editor in chief Kate is just off camera. Kate, things are going well. We have, we've got 2,300 people here. [25:51] So looking looking at cloud code together. Yep. Uh, [25:57] Oh, you want? Okay, let me just... [26:00] Thank you. [26:00] Yeah, maybe I should try that. If anyone from Anthropic wants to come on the stream and talk to us, I can see you commenting in the chat. I'll send you a link to StreamYard. Just like say yes, I would like to come on the stream and chat. We're very, very friendly. And I think a lot of people would love to hear from you if you're already here. And I will update the app. This is yeah, this is actually this is important. I have a beta build right now. So this is something that was not

[26:27] I didn't, I haven't updated my app and I'm a little afraid to do it on a live stream. Kieran, do you have it in your app? I'm curious. [26:33] I'm trying to get it working, but it was down for a while here, but it looks like some, it's back up again. Cool. [26:43] Anybody have other questions or things they want me to try? I'm taking requests. So ask for, ask any interesting queries. Most interesting query we'll put in every, when we do our vibe check. [26:55] Let's see. [26:57] I wonder if I could use this to code. I'm kind of like, let's see. Oh, it did our audit of [27:05] the Avery Agent Native Guide. [27:08] Can you see if it has artifacts as well? If it's similar maybe? Okay. It does have--well, it didn't do this one in an artifact, but we do have artifacts. [27:20] in another place. Let me just find it. [27:23] I do the fact that the context is clearly spelled out on the right. [27:28] Thank you. [27:29] Yeah. [27:30] That is nice. Context is here. [27:33] I saw the artifacts tab somewhere. [27:35] um yeah it's just like a friendlier it's like a friendlier version of cloud code to me but [27:41] Um, [27:43] Yeah, mine is working now as well. Can you find my every proof repo in Cascade projects? Basically, I want you to do a summary of the... [27:54] new feature, the provenance, the new feature provenance tracking that I've been building

[28:00] in Every Proof. [28:02] and write it up in a nice like [28:06] html file artifact that I can send to Kieran to explain to him [28:12] how the new provenance is going to work. I think this is a combination of something that [28:16] is kind of it's dev work related but it's probably not something i would ask cloud code to do [28:21] Um, [28:24] I don't have access to your Mac's file system in this Linux VM environment. [28:29] Interesting. [28:30] So that's interesting because it definitely does have access. Let's see. One of the things that gets confusing about this, I guess, is when it's running on your computer versus when it's not. Like, I think it's sort of unintuitive to the average person probably that when you're using it in chat. [28:51] It is all online and when you're using it in here, it's actually on your own computer. [28:57] I'm very curious, like [29:00] how they thought about making that clear from a UX perspective. And if anyone is from Anthropics on this stream, [29:07] Why does it think that it can't access my Mac's file system? [29:11] Oh, maybe I have to add that folder specifically. [29:15] Oh, yeah, that's what it is. That's so interesting. I just want I just wanted to YOLO give it YOLO access my file system to be honest with you. Look at all these projects by the way. This is how you know like I have a problem like these are all vibe coded projects basically. Let's see every proof.

[29:36] You know it's this one. [29:38] Okay, always allow. [29:40] All right. And now I can, the nice thing about monologue. So monologue is one of our apps at every is I can just, uh, [29:47] There's a shortcut for this. I can just click it and then repaste. [29:52] and [29:54] But yeah, what I really want, I just wanted to access my whole computer. It has that, that's interesting, it has that file cleaning prompt. I wonder how that works if it can't access. Yeah, I'm testing it now as well. I'm trying to... [30:11] see if I can use the scales I already have. [30:17] Oh, interesting. So I said, organize and tight up my downloads. And then it seems like it figured out this is so agent native, it figured out how to select the [30:25] It's as if I selected that folder in the [30:29] Mm-hmm. [30:30] in the UI, just because I specifically asked for it. That's cool. I think that's a really smart affordance. It just like... [30:39] I wish I had gotten... [30:41] activated here too. Ideally, it [30:44] It knows that I'm trying to access a folder in Cascade projects and it is as if I clicked this. [30:50] Are you sure? [30:55] It's the natural version. See if this is actually working. [31:01] Um... [31:02] Thank you. [31:03] Felix, are you joining? Amazing.

[31:07] Let me just find you on the... [31:12] X app, X the everything app. [31:15] And... [31:16] then if you can share my screen and I can show I will do that yeah let me share while you do that actually this is helpful all right remove [31:27] There you go. You're off to the races. [31:29] So I was trying this out. Help me generate a VST plugin, ask me user questions. [31:36] And I found a little bit [31:38] So I found some things. So basically the ask user question I love because it's this UI. It like runs you through and you can hit one, two, three, four, five, which is very unique. [31:49] Nice. The weird thing is... [31:53] I didn't answer it and it started automatically skipping this. [31:59] Maybe it's fixed now, but in the other one, it started automatically. Oh yeah, here, skipping. There we go. [32:05] See, so if your mouse is not on here, [32:08] It just thinks, "Oh, this user is not here, so we're going to skip this altogether." [32:14] which is very confusing. But also I love it because I'm a dangerously skip permissions person and I understand. [32:21] It's weird because if you're here... [32:24] scrolled all the way up, it will skip to the next. So just leave it. [32:29] strange here. But the cool part is here, I can say three [32:36] and it will go and continue. So there's like, I like that it keeps going and it's set to like keep going and finish. The skip UI here is a little bit weird, but let's say multi-type delay. And you can see it's working with my skill.

[32:56] Or... [32:56] Skills, skills, juice. [32:59] Oh yeah, it is mine. Happy Sharp Hopper is just the local place where this is happening. Let's say this... [33:08] So this is an interface that never existed before, which is cool. [33:14] Yeah, it's a little bit weird because it's in line here, but in reality you're answering. So... [33:21] I'm sure... [33:23] Let's see, sending requests. Yeah, so the skipping part is very confusing. [33:28] I don't know why it's skipping now. [33:32] But I do like, like, I would rather say [33:36] Maybe when I start the session, what kind of session it is like, if it's like YOLO, let's go session or. [33:43] where it like pings me and like very clearly in the co-work tab says like, "Hey, you need to answer a question here. I need your attention." Because [33:54] like there's something to be said to both and now it's kind of somewhere in the mail where it's not super clear. So I'd rather have it say, yell at me and say, "Yo, I need your input on something." [34:07] I don't want to give input on creating or using things, but if it will change the direction of [34:14] what this will be with the Ask User Question, probably that is handy to have. So far, this. [34:21] Interesting. I have a, since we have, it seems like there are some anthropic people on here. I do have a feeling about

[34:29] I'm curious what you think, Kieran. I just... [34:34] There's a limit to how many characters it displays. [34:36] And then it just goes over and hides the rest of my answer. And that just annoys the shit out of me. Do you know what I'm talking about? Yeah, I know. Yeah, it needs to be a little bit more flexible. Also, like, I want, like, why not 20 options? Like, why is five the maximum or something? Like, sometimes you just have 20 that you need to, like, I get it. [34:56] But also, yeah, stop questions, go build. [35:03] So that is nice. [35:05] just make it now. Okay, so it did not really [35:08] Like it's still doing stuff there, but here, like it's, I mean, it's a little bit wonky still. But I love the Ask User Questions flow. It's very useful. [35:21] Good evening. [35:22] Okay, so it does it here and you see the to-do right, which is here on the right. [35:26] Wait, I was distracted. What plugin are you building here? [35:29] Can you back me up? I think about a thousand. Yeah. Okay. So I have a skill called the juice skill, which knows everything about VST development. And I asked, can you help me brainstorm a VST plugin? And we're doing a delay. What's VST? [35:48] Oh, like it's an audio effect. So digital sound processing that you use in your music making. Yeah, yeah. And I'm making a delay now and it's building the delay. And like normally, normally the building of the delay, closed code is great for, but sometimes you want to brainstorm. You don't want to build. And that's why I like co-work because I just want to brainstorm a little bit. And the cool part is it has these skills. I see.

[36:16] So, yeah. I want to interrupt you really quick, Kieran, because we have a member of the team from Anthropic here on the stream. Felix, welcome. [36:26] hi friends how are y'all hey and how are you [36:29] Uh, we never met before. Tell me about you. What do you like? What do you do at Anthropic? Like, how are you involved in this? [36:35] How am I involved in this? I've been in the topic for a little bit, but this is the product that my team has built here. We sprinted at this for the last week and a half. What we're trying to do here... The last week and a half? That's it? Come on. To be clear, I think many people have... [36:53] had the idea that something like Cloud Code for non-coding work [36:57] would be helpful and useful to people. And fundamentally, what we want to do here is we do want to help people out with their work, like whether that's a personal thing or a corporate thing. [37:05] And we've had a different number of prototypes [37:09] in particular before Christmas, [37:11] But I think [37:13] Over the holidays, one thing we have seen, I'm sure many people have seen this, is that an increasing number of people is using cloud code for almost anything. [37:20] just like we are, right? We're sort of like automating our entire lives with cloud code. So we were thinking, what is a small early thing that we can try out and ship to people [37:29] and iterate with them together to really figure out what is the right [37:32] What is the right user experience? What is the right thing to build? And this is it. This is the sort of like research preview, very early alpha. [37:42] a lot of like [37:43] a lot of rough edges as you've already seen. There's a lot of things about it that I think we're going to improve very quickly.

[37:49] But this is our attempt to build in the open and work together with people out there. [37:54] I love it. Tell us about like some of the design decisions you made, like an early one, for example, is there's a third tab instead of maybe adding a co-work mode into the chat tab. Like, how did you think about [38:06] and what was the process to come to the design that you have currently for how the product works? [38:13] That's a great question. [38:15] I think one belief I have is that [38:19] The current user interface that you see across agentic applications, not just in Anthropic, but across the industry, is probably going to change pretty dramatically in about a year or two. Right now we have these hyper specialized individual input fields and we have a lot of custom scaffolding around the specific tasks that you're going to do. But as we see the intelligence of models. [38:38] improve and as we also like maybe holistically as an industry figure out a little bit of the generalization problem i expect that we're actually going to see a smaller number of interfaces for a wider range of use cases for now what we're doing is the reason we broke it out is because we want to be pretty transparent that this separate thing is a construction site right that we're sort of letting you into our kitchen we want to like work together with you we want to ship almost every single day some new features some bug fixes try out some things so this [39:08] per tab is fairly experimental. [39:10] It's [39:11] you could say on the frontier or the bleeding edge, but it's just a little bit less polished and a little faster pace. [39:17] And that's one of the main reasons why in a separate tab. There are some technical reasons too. I could tell you like one of them is that currently this is running on your computer. So your chats are local. They're not shared with other devices. We're being a little bit more aggressive in how many agent capabilities we give cloud.

[39:34] Those are the main reasons. [39:35] How did you think about, because I feel like that's such a huge UX hurdle to get over. How did you think about... [39:41] letting people know, hey, this is actually running on your computer versus chat, which isn't the same application is not. [39:47] Yeah, that seems so hard. [39:50] Yeah. I think the dream that I have, and I'm sure many people have this dream, the dream that I have is that it doesn't really matter. Where your code runs, it should be technical implementation detail, and it should matter to people as much as when you visit the New York Times.com, like is it using WebSockets or not? [40:05] And it's like, [40:06] Who cares? I think for us right now, it's an opportunity to move a little bit faster. [40:12] and to ship a little bit quicker and also like work a little bit closer with the people for whom we're building this i have this strong belief that it's very hard to figure out a great product in isolation by yourself right you sort of like go up into a cave and you work on something for a year and eventually comes out i think it's really hard to build a good product that way um and i often like to remind people that like even the first iphone was missing a bunch of things that [40:42] But we're okay with that for now because we do want people [40:47] who are signing up for this ride to sign up for it fairly intentionally. [40:51] Hmm. Yeah, I think that's a really interesting pattern is like [40:55] Let's ship really fast and we'll ship it as a new thing in the app that maybe fewer people will click on. [41:00] um so that we can get it out in the open and start iterating together rather than like try to make it perfect especially in this world where it says you said you were working on this for a week and a half this version for a week and a half which is

[41:11] kind of insane. Um, [41:14] Hmm. Kieran, do you have any questions? [41:16] Yeah, I'm curious. [41:18] Clearly, this is the version that's out now, but what is the version in your head? Where do you want to go next? [41:32] things you're dreaming about. You used the word dream. What are those things where you want to go? Because I'm sure everyone on the team had wild ideas and then they were like, no, we need to ship Monday, so let's just like... [41:46] If you can share any of those, we'd love to hear those. [41:50] I love that question so much because I think I actually have the same question for the two of you, right? Which is, where do you want this to go? What do you want to do? I've already heard you say you kind of want to give it to access to your entire computer, the multiple choice thing where like actually can we like shift around a little bit and how we want to do this. I think right now, I am much more in a mode of, okay, let's see what people think and then try out a [42:12] billion things. Some of them will probably be the wrong thing. Some of them will be the right thing. But I think it's much more interesting to me what people want to do with this rather than like what's my own personal dream or vision. That makes sense. [42:24] in the things that I've built in the past, this was always the thing that happened. You have an idea of how people will use the thing that you build. They actually find a use for it in all these other ways, and then you lean into that. I'm really hoping that [42:38] I'm really hoping that we can learn a lot about what do people want, what do people not want, what do they like, what do they dislike. I'm sure people will dislike a few things about this and then we adjust and

[42:47] iterate on it. [42:48] That's the really cool thing about... I'll go for it, Kieran. [42:51] Yeah, so Boris is very good in building cloud code in a way that people can figure out what they want. Like, is there a way, like, do you use that strategy in a way as well where you give some building blocks or things for us? Like, for example, can I include my own plugins or skills? Is there a way for people to experiment inside Cowork as well in maybe the non-coding way? [43:21] Is it really like this is the product? It's that's what it is like how because there's [43:27] cool balance between how Cloud Code works and people that use it, because it's super hackable. Is there a similar philosophy in Co-Work as well for non-coders? Yeah, very composable. The first thing you said about Boris being very good at steering Cloud Code in this direction of shipping early, and then iterating on it and seeing how people use it. It's really funny that you mentioned that, because I think one of the reasons we've shipped today, maybe shipped a little earlier, [43:54] Well, it's because Boris pushed me and was like, hey, [43:56] You should probably show this to people, see what they did. On the composable piece, I think the [44:04] thing that I found most impressive in my own work over the last couple of weeks and [44:08] maybe sort of the last two months, is that I'm really leaning into skills, [44:12] Instead of previously writing MCP tools, this very specific harness that is very tailored towards just Claude, I instead just write skills. Sometimes I still write a binary and then I describe in a skill how to do something. What's a good example? I'm working on a marathon training plan for myself. I wrote a little binary that fetches all my athletic activities from various pages.

[44:36] But then I just write in Markdown in a skill file, hey, Claude, if you want to make a training plan, please follow the following guidelines. We do automatically load any skill you have installed in Claude AI into CoWork. [44:48] And I think that's probably going to be increasing, especially as well as it's and especially with Opus 4.5. [44:54] It is so good at following skills. [44:56] So, skills is probably the primary hackable surface. [45:00] that I'm exploiting right now. That's great. [45:04] One thing you said earlier in the conversation is you think that there's going to be fewer [45:09] like UI services? Does that mean that like over time there'll be fewer UI services? Does that mean, because there's a lot of debate over the last couple of years about is chat the final form factor for AI and everyone's like, no, we need more UI. Are you, is that [45:22] Are you putting your stake in the ground as [45:25] natural language actually is here to stay. And we're going to have fewer UI services where you just talk to [45:30] an agent, maybe an agent orchestrator that goes and [45:32] talks to a bunch of other agents. And that's the kind of form factor you're pushing towards. So it looks a little bit like how cloud code does today. [45:41] Yeah, I think this is still very heavily debated. And there's certainly no anthropic viewpoint. I'm not even sure that there's a viewpoint that my fairly small team would holistically agree with. I think people have very different visions about [45:53] How will people interact with [45:55] AI and models in the future. If you ask me very personally, I think I believe two things. One is that [46:02] the chat input and its various forms, not just for models, but in general, right? Like the idea of this text box and you put into the text box where you want. If you generalize it enough to say even Google.com or the

[46:14] address bar in Chrome is like a I want something input box. I think that is going to stick around for much longer than we all think. [46:22] That's the first thing I think. I think there's [46:24] we will continue to have something that looks like a search I want something box. [46:30] The second question is how many separate boxes do you have? Do you have one box for code? Do you have another box for [46:37] maybe like a personal entertainment, you have another box for healthcare related concerns. I'm not sure we're going to have too many boxes of those. And there too, maybe I would go back to Google [46:46] I think I remember the early 2000s where you had a different search box for every single Google sub product. Increasingly, you just type what you want into your Chrome search bar. [46:57] And you don't actually go to sub page of sub page. I'm in the mode right now of looking specifically for shopping things. So I go to Google Shopping. I would be surprised if we don't see a similar [47:10] generalization that is smarter about figuring out what you want to do [47:13] in the future. We might still have different interfaces where it sort of like splits out, right? And understand that you're trying to do X, therefore I'm going to show you UI for X. But the entrance point, [47:24] I think, yeah. [47:26] The interesting counterpoint to that is something like Microsoft Excel, which I think it also has some similarities to the way that... [47:33] you know, just generally AI works is it's this general purpose product. It's super simple to get started. You can make things really like endlessly complex with Excel and then Excel sort of spawned this the B2B SaaS wave like you probably don't get B2B SaaS without Excel. So there's I think there's also the other argument that you have these sort of general

[47:52] really general tools and then people find power workflows within them that then get split out. [47:58] Yeah, yeah. I think Excel is such a beautiful example of so many things because it's like, for many developers, something that sort of exists a little bit on the periphery, right? [48:07] And [48:08] I've often heard the analogies between how many daily active users Excel has versus how many developers even exist on the planet. It's an interesting number. And I think [48:19] i think [48:20] The thing that I find interesting about Excel and the [48:23] the commitment it has from its power users, is that those power users are not too interested in [48:33] marginal productivity gains or marginal UI or UX gains [48:37] over deep familiarity with the product. [48:40] And I think that's interesting. I think there's a lesson there. [48:43] in some shape or form. And I think I've actually seen that across various other surfaces where [48:48] You as a developer sometimes look at someone's workflow and you say, oh, I can make this workflow slightly better for you if I make you a specific. [48:55] use case tool over here on the side. And then people sort of fail to adopt that thing because they're actually more comfortable doing specific things within their product. [49:05] Um, [49:06] As an example, I think that's a lesson that I have learned. I was previously at Slack for many years. [49:11] And there's a lesson that I've learned there over and over again is that you can make these separate services that you think might serve people's use cases much better, but they will continue to just do it in chat. Hmm. [49:21] That's a really, really good lesson. I love that. Speaking of speaking of that, I think there's

[49:28] Today is for the non-developers, but I feel like there's a lot of developers who are watching this right now. And you're someone who's... [49:36] if you built this, so like you're deep into how to build like agent native applications. This thing that we've been thinking about and talking about a lot at every, we just published a guide called agent native architectures. And we've been thinking about like, [49:50] what are the core principles of agent native apps? And I'm really curious if these resonate with you, if you think they're wrong or if there are things that you would add that are part of how you guys at Anthropic [50:01] think about building agents. So an example is [50:04] So one of the things that we think about when we build agents internally at Every is, [50:09] Whatever the user can do through the UI, the agent should be able to do. [50:12] And I see that a little bit. That's basically how cloud code works. But I see that a little bit in what you built with co-work where, for example, if you didn't pick the file picker, it'll automatically determine that you are asking it to pick a particular folder and it will do that for you without you having to touch the UI. So that would be an example of parity. [50:31] Another one is granularity, which is basically [50:35] Tools should be [50:37] mostly at a lower level than features and the feature should live in the prompt or the skill so that you can combine tools in new ways that you didn't predict previously. And then that allows for the third one, which is composability, which is you can combine, you can combine those new ways and you get

[50:59] the fourth one, which is emergent capabilities. [51:01] people are just doing it for stuff that you didn't expect. And you met you see the latent demand and then you like build for that. And this is essentially I think like a lot of my summary of how cloud code works. I'm curious of like how this sounds to you and if you think that they're we're missing anything or are there any things that you've learned from doing this in production at a huge scale that could make it make people better at building these kinds of applications. [51:26] I think this really resonates with me, right? And I think one thing that's hidden in emerging capabilities is the [51:32] the inability I think especially individuals and silo teams have to predict [51:37] how an agent actually ends up being super useful if you give it fairly primitive tools. [51:42] I think pushing down tools into like a general space [51:46] is very powerful. The more composable they become, the more generalizable [51:51] the more generalizable the tool is, the more you will benefit from improvements on model intelligence. And I think for [51:59] Many developers that I've been talking to in the past [52:02] It seems like the rate at which model intelligence and models ability to call tools effectively improves is actually much faster than your ability to like maybe churn out additional tools and like educate users on them. And I think if you, if you sort of take a step back and you think of how can I build a very generalizable tool, you have a much better chance to build something that can adapt to new use cases. [52:24] So I think that resonates with me quite a bit. [52:26] Mm-hmm. [52:27] What about the like trade offs? Like I've been talking to Kieran about the trade offs in tools. Kieran, do you want to talk about the what you're kind of what you notice in Quora and what you're thinking about?

[52:36] Yeah, so... [52:38] I think putting things in a prompt... [52:43] is great and then having the tools, but there's like, we need to now suddenly create tools that then [52:50] read skills or something like that. So like we have to invent this meta layer, like skills is like just in time prompt injection, but like, [52:59] We need to create that thing. And now everyone that's building stuff, unless you use Cloud Code or the Cloud SDK, it's all built. But... [53:07] Like, it's this thing, like now there's like this struggle of like, oh, but like, [53:12] tools are that you can describe stuff in a tool or you [53:16] create a tool that then wraps around it and then calls something else. So there's like this friction there and [53:23] like it is great to make things composable. Like if originally you create, for example, like five tool calls, one a search email, like read email, this and this and this. But you can also say, no, we do just do like an execute tool call and we create skills that can do those things or an MCP or some obstruction there. So there's like this change happening. And obviously this is like the clause code is like, and the clause SDK is a very good [53:53] push for that. But I feel friction there. I'm sure you felt that friction too. So maybe you have some best practices for people that are stuck in the old fashioned AI world and need to go to the more agent native things that you learned or that you've noticed because you've implemented on top of, I assume, Cloud SDK, maybe, or some variant. So you use that and you implement things on top.

[54:23] air [54:24] I'm not sure that I have any wisdom from the mountain that is going to be more valuable than yours. [54:29] But I think what you're saying... [54:32] I think what you're saying that resonates with me is that [54:37] you sort of need to make a call, right? Like which part of the outputs do you want to be non-deterministic and where are you comfortable, where you're comfortable relying on model intelligence. And every single time you do rely on model intelligence, if you pick a cheaper model or like a dumber model, then those parts also go down in quality. And I like to break up my workflows into like the non-deterministic and the repeatable parts. I think the more repeatable something is and the more easily I can say this will never change. [55:07] I think that's a place where [55:11] where it might make sense to write a tool. And in a sense, we're already doing that. We're not implementing [55:17] You could give Claude a very generalizable [55:21] write assembly code tool, right? We could be like, just call GCC and figure out whatever you want. But we don't, right? We give it. Yes. [55:31] That's Dan's ideal. That's the most granular you can get. [55:37] But I will say that I think when I talk to developers out there, depending on how... [55:44] sci-fi you are i think even that assumption is a little bit [55:49] I wouldn't bet too much money on it. That assumption is certainly under attack. Like the idea of like, should you give cloud any tool at all, or should it just be like a.

[55:57] Claudia's memory, start writing ones and zeros, go wild. [56:01] It's an interesting area. It's like hard for me to know right now. [56:06] - No one knows, but you learn stuff. You created skills for exactly this reason. [56:12] Because you need it more than just a slash command and a subagent. And you're like, yeah, like we need the Cloud MD to be better. But like, I guess that's why skills... [56:26] like were created and clearly that's working well. I resonate with you saying like, skills are amazing. This is also like, I'm creating skills every day and [56:35] I love them. So clearly there's something there. But when do you like, [56:40] When when [56:42] do you not have it be a skill? It's very interesting to me. You know, I think this is such a fun conversation. Someone you should actually talk to at some point is Barry. [56:53] Um, [56:54] because Barry is the one who [56:56] is [56:57] at least inside the company, sort of like our skills person is the person who essentially came up with skills. And for us, fundamentally, skills were a little bit of a byproduct of the same tension that you're describing. So what we wanted to do is we wanted to make a very easy way for... [57:12] people inside the company to get dashboards. Right. And we have a, we use one of the popular data providers where we keep a lot of our data. And we were trying to figure out, okay, do we build [57:24] really specific tools that fetch that data and then compress it down into like a specific format. The first couple of dashboards, Claudia, in the building looked a little

[57:33] you know, this was before 4.5, the dashboards were not ideal. Every third or fourth dashboard it generated was like a little leg luster. So we did think about, okay, do we like super [57:45] parameterize it and basically build like a fixed dashboard that you cloud and only plugs new data into but in that process while building that we sort of discovered hey if you just tell claude [57:56] how to effectively query this data source. [57:58] that it can use SQL and that it please follows the following design guidelines for making dashboards. Suddenly you get something very, very good and you get it very good every single time. And then you also give people, and this is the emerging capabilities part, then you also give people the opportunity to say, "Hey Claude, [58:14] I understand that you're following these principles for dashboards, but I also want [58:17] I don't know, different chart type, or I want to combine it with other data. And that's then where things really open up. [58:23] Right. [58:25] That is really interesting. I feel like the one one way to talk about why you might want to skill instead of just having it have GCC and just everything is just in time is it's like about sharing something repeatable with other people that you can talk about. [58:40] And there is something actually about not everything should be just in time, because you want to do the same thing over time with a group of people. And that's that's kind of a skill, I guess. [58:51] Yeah, that's sort of how we operate too as humans, right? Like when I join a company, someone tells me how to book a flight. Like, how do you get a room? [59:01] I think a lot of us operate even as humans on a long list of Markdown files.

[59:06] with stuff in there. Felix, I want to give you an option. You've been very generous with your time. I want to give you an option to hop off if you want. We would love to keep chatting. We have endless questions, but I'm sure you have a lot going on. Do you need to go or do you want to keep chatting? [59:21] I think this is a good time for me to bounce, but I'm not going to go before both of you give me one thing you would like us to change. [59:27] I mean, my easy one is just YOLO access to my whole computer. [59:31] Okay. Okay. And, and make it easier for me to know whether it's working on my computer or working in the cloud on a chat and make it easy for me to use it on mobile. [59:43] Okay. [59:44] Yeah, plus one on the mobile, but my favorite thing would be [59:50] the ability to add my plugins. So just my, I have a marketplace with plugins. I just want to hook it up to GitHub. Fair enough. Yeah. Yeah. Like, because now I'm like adding things in the app and then copying it there. And like, it's like, I like it. Probably I can just copy it somewhere, but like just native support for a marketplace and then adding it and syncing it. That would be absolutely great. [1:00:13] Thank you. I appreciate both of those. [1:00:15] quite a bit. We're going to take those back, going to tell the team about it, and for everyone else on the chat, find us on the internet, send us what you think. [1:00:22] but quite interested in hearing from people. [1:00:24] and adjusting our roadmap. [1:00:26] Thank you so much, Felix. Thanks for building this. Thanks for joining. [1:00:30] Thank you. [1:00:31] Have a good one. All right. That was awesome. So cool. [1:00:35] Thank you. Yeah. And we have so many people on this stream. There's almost 10,000 people here. Oh my God. If you're joining us for the first time, we just had Felix on Felix is a member of the technical technical staff at Anthropic and he was talking to us about Claude Cowork. If you are looking at this, this stream, then you probably know what Claude Cowork is, but I will share my screen and show you again in case you're.

[1:01:05] of Claude. That's sort of like Claude code for non-technical people. It looks a little bit like this. We've been testing it at Every. We're the only subscription you need to stay at the edge of AI, Every.to. We get access to this stuff before it comes out. And we do vibe checks like this. We do them live. We also write them on Every.to. So we got this earlier today. So we were just testing it out [1:01:28] Here's an example of a task I gave it. And you'll notice it looks a lot like [1:01:33] normal Claude, like the normal Claude chat. The differences are a bit subtle, but I think that they [1:01:40] it does make a big difference. So instead of a chat, you have tasks. When you look at the, you know, for this query, [1:01:48] Or maybe like, let me find a better example of this. Like this query, for example. You'll see that co-work ran for a really long time. If I ask the same query, I want you to go to our every agent native guide and walk through it to do a UX review. [1:02:06] Claude would have stopped after a couple turns and given me a pretty good answer, but this does a lot more research. So because it's working on your computer, it's going to be able to work for a long time. And I think that's sort of the key thing that you can take away from when you might want to use Co-Work and how Co-Work might work differently than regular Claude, which is [1:02:24] For the first time, if you're non-technical, you can ask... [1:02:27] your computer to do something and walk away for a while. And then you can also ask it to do many things in parallel. So this is an experience that if you're using cloud code, you have a lot with programming, but I think most non-technical people are still in the kind of like turn by turn era of using chat and just expecting a response almost immediately. And this is much more of an

[1:02:49] built to be an async experience for non-coding tasks like data analysis or research or writing documents, all that kind of stuff. [1:03:00] Like Felix said, they built this in a week and a half, which is crazy. Crazy. Which is the new normal, by the way. So anyone who's like coming up with a PRD in two weeks, nope, you ship the whole thing in a week and a half. [1:03:15] 100%. So there are some rough edges here, but they're going to be improving them really quickly. I think it's really cool. The pattern that he shared that I really think is interesting is I was kind of asking why even add a new tab, like a co-work tab. And he said, we essentially needed a playground to mess around and do stuff that is a little bit less polished. And so it's nice to have this extra tab, which is an interesting pattern in AI where now you can build so quickly. I think we need more patterns for what to do with that. [1:03:45] So, Kieran, any reflections from our conversation with him or anything you're thinking about right now? [1:03:51] Yeah, I think it's like why use co-work over chat or code. [1:03:57] That is always my first question. And I think if you're a non-technical user, [1:04:01] just think of Cowork as something that is chat. Just try Cowork as the new version of chat, the better version or a different flavor. And just open it and do the same things you've been doing in chat. But [1:04:15] you see that it is different and you can do things. Like one really, really important thing is in chat if it

[1:04:21] is responding, you cannot send a new message. When it does something, you're like, "Oh, wait, wait, that's wrong." With Co-Work, you can queue a message and [1:04:30] like while it's working, say something new. So like just those tiny things are very, very handy. So just try co-work instead of chat. It has skills like you can generate documents, Excel sheets, PowerPoint presentations, PDFs, like it can do all those things. So [1:04:47] Um, [1:04:48] if you are applying to jobs, like just upload everything and say, hey, can you rewrite my resume for this job? [1:04:55] job and try it out, things you would manually do normally. [1:05:02] And also, connect Chrome. You can enable it. Maybe you can share, you can show that how to do that then in connections. Yeah, so if you go into your settings and go into, I guess, is it connectors or is it extension? It's in connectors? Yeah, control Chrome. So you can add that in there and then it'll just be able to basically use Chrome on your computer, which is really cool. I mean, one of the things I had to do. [1:05:29] that is totally new and interesting is I have it I had it go through my my Twitter feed my ex feed [1:05:37] And I asked it to just like, tell me what was hot right now. What are people talking about? [1:05:43] and that's really cool like that's that's ordinarily something that you would have had to [1:05:47] pay a lot of money for an API for, and this can just do that without really a problem, which I love. I love that. And if you use Chrome and you're logged in,

[1:05:59] on things in Chrome. It's already there. So you don't need to log in again. It just uses your Chrome itself. Yeah, exactly. And I use Atlas, so I had to log in all this stuff. But if you use Chrome, it's great. So like, look, it just read my Twitter feed. That's so cool. There's so many possibilities if you're a writer or a marketer or... [1:06:21] anyone that does research, especially research on things that don't have APIs, that you can now do without much trouble. And to your point earlier, Kieran, like when you're thinking about, okay, what would I even use this for? [1:06:34] I think that one of the things that we try to embody at every is [1:06:38] We try to be really curious. [1:06:40] and know that [1:06:42] If you're trying new technology for the first time, the first five things you do probably aren't going to work. [1:06:47] But there's like a really, there's something really fun about being curious right now because Felix, for example, the guy who made this doesn't even know how we're going to use it. [1:06:56] he needs people like us, anyone who's watching the stream to like, [1:07:01] mess around with it and figure out all the new, interesting, emergent ways that it could be used so that he can make the product better. So it's the first time where the software developer is more like setting up a playground, but has no idea how the playground is going to be used. And the users are the ones that are being creative and figuring out [1:07:19] things to do. [1:07:20] And so if you've [1:07:22] if you approach this with curiosity, it's like a huge opportunity over the next couple of weeks to like figure out what this is for.

[1:07:30] I think another important thing is people tend to have this thing that I like to call capability blindness, which is I tried this once before three months ago. It's never going to work with AI. [1:07:41] And the really interesting thing about AI is like, it changes every couple months. Like I, [1:07:48] freaking like Opus four or five just totally changed everything. Stuff that had never worked before starts started working now. And so [1:07:57] If there's something that you really want AI to do, like for me, that has always been, I wanted to do copy edits. Oh, we should see if it can do copy edits. I'm going to set that up. [1:08:07] So I've always wanted to do copy edits. [1:08:09] Thank you. [1:08:10] And every time something new comes out, I just try it because I know at some point in the next couple months, it's going to start working. So I'm going to actually set that up. I think that'll be a really fun demo. Kieran, do you want to talk about anything on your mind so that I can take a little bit of time and get the copy edit set up? Yes. Yes. [1:08:29] I'll share my... [1:08:32] Here, I'll just share a screen here. [1:08:35] So, [1:08:36] Yeah, if you share my screen. Oh, sorry. Yeah. Forgot that you were, that I was, you could see my screen. Okay. We're professionals, everyone. Yes, obviously. We started live streaming last week. You're right. Give us some slack. Yeah. Yeah. [1:08:52] Cool, yeah. So we talked about skills a lot. You might think like, what the hell is a skill? Isn't that just a prompt? Yeah, it's a prompt, but it's also more, it's more like,

[1:09:04] Yeah, and what can you do with it? [1:09:07] This is the most hackable way for co-work. So if you want to personalize your co-work experience, this is the way Anthropix says. So I asked what are all the skills I have and how can I use them? So for example, these are document creation skills. [1:09:22] And it just learns how to create an Excel sheet. And these are some of my own. I love Swiss design. So I have like a Swiss design scale, a Gemini image gen, where you can get nano banana images inside code code. I have a DHH Ruby style to roast my code. [1:09:46] Yeah, all these things. [1:09:48] I just create skills for everything. So I have a 3D print skill where I needed to print some 3D things. And I was like, I'm sure Clouds Code can do this. And so I created the skill. And these skills, normally what I do is I go to chat and like say, [1:10:10] can you generate, can you deep research how dense shipper rides? So for example, if I like dense riding, I might [1:10:20] Do this and actually... I don't know. That guy's kind of a blowhard. [1:10:24] I know. So normally what I do is I enable deep research here and [1:10:30] go hard and it runs deep research runs for like an hour sometimes which is great and then I say I

[1:10:37] can you create a skill for this? So I'll just fake do this, but [1:10:42] Then I create a skill for this and there is this thing in called "im". [1:10:48] capabilities that you can enable, which is really cool, the example skills. So you go to capabilities, example skills, and there's a skill creator from Anthropic. If you enable this, after you did deep research or anything, you can say, create a skill out of this. Obviously, you can do that here as well. So you can do research in Cowork and create a skill out of this, and it will be loaded inside the [1:11:14] inside here. So for example now [1:11:19] KU design with Swiss design a very beautiful chair for me and create a STL so that I can 3D print this miniature. [1:11:29] four centimeters high. So let's see. So I do this and I shoot them [1:11:33] And [1:11:34] KU Design. [1:11:37] Okay, so hopefully it's loading the two scales here. You can see the context it pulled in the Swiss design scale and the 3D print scale. So I like that, that you can really clearly see this. [1:11:52] And what it does, the skill will just inject a prompt where it says, just do these things, make it look good. Don't use inter or whatever. This is the aesthetics. So it's now doing things. And there is also in skills, you can put like scripts, Python scripts, whatever script you want, binary scripts, anything you want to run.

[1:12:22] script and put it in a scale so every time it runs. For example, you want to check if it's the shape or if it's following a certain [1:12:30] like static or if it's actually linting well, you can encapsulate all that in the skill and trigger it as well. [1:12:39] So [1:12:41] It's now doing this, which is really cool. And the scale will create a STL file, which I can then 3D print. So we'll see. Like a chair, obviously, is maybe a little bit hard, but... [1:12:53] Why not? So this is how you can create your version of co-work [1:12:59] with skills. So you can capture and find the things you do and encapsulate your style and everything you do into a skill and then have it available here. What is cool here is on the right side you can see the STLs out the preview. There's a preview. Let's look at the preview. [1:13:20] Okay, so it's a little bit hacky still because, or here the side, there we go. Okay, this is the chair, beautiful. So this is SVG, it's not 3D. Let's see, can we look at this? No, we cannot look at this. [1:13:35] But yeah, we have a STL. Let's see if I can open this. Open in Bamboo Studio. I'll see if this looks any good. [1:13:46] I'll share my other screen. [1:13:48] Share screen and then we go to Dan

[1:13:53] So here is my Swiss design chair. - Amazing. We need the skill, Kieran. [1:14:03] Oh, yeah. So actually the 3D scale is on my machine, but I'll share it. I'll push it to the interwebs. And that was actually my request to Felix. It's like, can I automatically pull these things into my coding experience or in co-work if I push them online? [1:14:33] go wild. And it's kind of funny to then print this. I'll print it and take a picture later. I love it. So I, if you've just joined, one of the things that we're talking about with Claude Cowork, I think that one of my, [1:14:48] bars for AGI is can it do copy edits in a Google Doc? Like it's surprisingly hard to get these things to do that. So what we have at every, obviously we publish every single day. We have a very high bar for the copy that we publish. We want to make sure everything is like really, really clean and beautiful. And so we have a skill that's in Claude that is our every proofreader skill. [1:15:11] And what I wanted to do is see, can Claude co-work copy edit a Google Doc? So this is the Google Doc. This is an article we published last week by Katie Parrott, who's one of our writers, who's fantastic. She's going to be the one writing the vibe check on co-work for later today. So look out for that on every every.to.

[1:15:29] So I just said like, okay, I have an every proofreader file. I just downloaded the skill file and gave it gave it to it and documents. Can you go to this Google Doc? [1:15:37] and make edits as suggest changes. [1:15:41] and then uh this is actually really interesting so like it's just hard for ai to do this because [1:15:45] It has to go through the copy editor and just look for every, for each rule in the copy editor, it has to go through every part of the document and find all the violations and that's just like really hard to do. But you can see, I said, go do this. This is not something that I don't think that the regular cloud app could really do this. Maybe cloud if using it, Chrome could. But other than that, it could not do it. You can see it loaded the document. [1:16:09] um it's getting all the text it's scrolled through the whole document [1:16:13] And it's now clicking on, it looks like it's clicking on suggesting. Let me see if I could actually find the Chrome tab that it has opened. Yeah. So you can see that it's. [1:16:23] Here's what it's doing. It is [1:16:28] It successfully got itself into suggesting mode. And now what it looks like is... [1:16:34] It's. [1:16:35] searching using find and replace for errors that it found and we'll see [1:16:41] Now I need to double click on Editor in Chief to select it and type the replacement. Let me triple click near it to select and manually make the edit. [1:16:48] This is so interesting. [1:16:50] um i feel like watching this is like watching a video game i'm like oh yeah come on come on claude yeah it's really entertaining yeah yeah but google google docs is like the final boss of stuff for ais to use because it's just like so it's such ultimately such a simple application but it's the way that they built it is so complicated it's like not actually

[1:17:15] real HTML. It's like a whole [1:17:17] It's just just. [1:17:18] really hard. So, um, yeah, it's just struggling. So if you're at Anthropica and you're watching this, if you could please improve your computer use to actually be able to use Google docs, well, that would like totally change my life and change the life of Kate, our editor in chief and a bunch of other people here. So, um, [1:17:34] Yeah, please do that. [1:17:39] Everything about AI is like a video game. Absolutely true. Kieran, go for it. Yeah, so I'm thinking, so we have this skill, but is it like skills can be maybe optimized now to actually make use of sub-agents and things like that, that maybe were never available? So it might be also time to rewrite our skills a little bit. [1:17:58] if you created skills for Claude. Maybe... [1:18:04] we can push it to use sub-agents or you execute scripts more or like do some things in a programmatic way. So there might be an opportunity also [1:18:14] to do more of that. Yeah. [1:18:17] Katie Parrott, who wrote this article, says, surely this is the cleanest copy that is ever copied. It's true. It probably hasn't found many errors because Katie didn't make any. But we'll see. It's still working on Editor-in-Chief with dashes. It's doing something. Thank God this is able to work async. Because if I had to, like... We were actually... [1:18:38] looking over its shoulder for forever, it would not be particularly useful. Yeah, we'll do a live stream, or Claude should do live streams of it doing work. Yeah. Actually, that'd be really fun. Claude doing work dot TV. And then see it mess up.

[1:18:54] Here, I'm going to give it a hint. Hey, buddy, you can just type in there and replace it. You don't have to use find and replace. Let's see if it... So you saw, yeah. You saw Dan do that. [1:19:08] Yeah, this is another thing that this does that is different from if you're using the chat, which is you can see this button says Q add to Q. And this is going to be a lot. The UX of this is a lot more like the UX in Q2. [1:19:22] um cloud code where you can just send messages even if it's working and it'll deal with the messages [1:19:30] You don't have to wait for it to respond to your last one. And I think that's really that's really, really good and important for [1:19:36] async conversations and we got our first compact the the bane of every clod app users existence is compact um so we'll see coming up after the break we'll see you claud can replace editor-in-chief with dash or forgot what it was doing and does something completely different we'll see so one one thing it says q which is a little bit confusing because q in my [1:20:06] would mean like do this after you finish the one before. But it's actually a little bit smarter. If you say, stop, stop, stop what you're doing. This is terrible. Like it will actually look at the text and think like, hey, do I need to change anything what I'm doing right now? So it does like either pick it up immediately or if it's like after this, do this, it will actually cue it. So there is a little bit more flexible than just like,

[1:20:32] injecting it, it will look at it whenever you add it, which is good to understand as well. I wonder if Klaue is getting performance anxiety because it knows that 13,000 people are watching it. Okay, dude. Oh, I thought it was 13 people. [1:20:47] are watching this please just type you're so close um if you if you want more [1:20:57] extremely smart and insightful takes on AI, you should subscribe to Every. Every is the only subscription you need to stay at the edge of AI. It's every.to. [1:21:10] We we have a pretty cool business. We do. [1:21:14] Ideas apps and training at the edge of AI. So on the idea side, we have a daily newsletter. [1:21:18] We get our hands on stuff like Claude co-work early. On the day that these things come out, we do vibe checks, which tell you from our team as we're using these in our day-to-day life and work. What is it good for? What is not good for? That happens for products. It happens for new models. Like when Opus 4.5 came out, we had a review on the day of. I can show you that. We also develop apps ourselves. [1:21:48] I'll talk about the apps in a second, but this is the article that we wrote on the day that Opus 4.5 came out. This is by Katie. Again, this is Katie's article by Kieran and by me. And if you've been seeing all of the hype about Cloud Code and Opus 4.5,

[1:22:06] We started talking about it November 24, 2025, the day it came out, and we said it's the coding model we've been waiting for. It took people about a month to catch up to that. So if you really, really want to know what's going on at the Edge of AI, it's really good. I mean, I'm biased, but I think this is a really good read, and our vibe checks are really good. We also have apps. So we have four apps that we build. As part of every, we have one called Quora, which Kieran builds, which is an AI email assistant. [1:22:36] our computer. We have one called monologue, which is a speech to text app. You can see monologue right here. This is monologue. I'm talking and it will type in here. [1:22:44] I don't want to mess up our buddy Claude, but it's sort of like Whisperflow or SuperWhisper. [1:22:52] And it copied it to the keyboard. You see it. It just pasted my text in there. We've got a couple other ones. Spiral, which is an agentic ghostwriter, and Sparkle, which is a file. [1:23:03] manager, file cleaner. And it's all available for one subscription. So you pay one price and you get access to... [1:23:12] all the ideas, so everything that we write, all the apps, so four AI apps at the edge of AI, and training. So we have camps that you can go to. We're doing a cursor camp where the team from cursor is joining us in a couple days, and you get to learn directly from the cursor team. You saw we had Felix from Anthropicon earlier. We do these all the time. We basically teach you how we build. So everyone at Every is a builder and writer, and as we're using these tools like

[1:23:42] code to build apps and do writing and do design and all that kind of stuff. We bring you along for the ride. You should subscribe every.to. And we're still on this extremely... [1:23:54] scintillating view of Claude trying to make make suggested changes in a Google Doc. And I think we can call we can call this one just we need a copy edit Google Doc copy edit benchmark and it has it has failed that benchmark is sat is not saturated. [1:24:10] We're still not an AGI folks, so stay tuned for hopefully the next model. [1:24:16] Oh, Anthony, Anthony Morris from Anthropic, huge fan of Quora. I love to hear that. Kieran, does that make you feel good? Yes. Yeah, we have a few Anthropic peeps using it. That's pretty great. Thank you. [1:24:29] Kieran, what's on your mind? I think on my mind is I want to try this out. This makes me very excited. Like this feels like something that was missing. Even as a coder, I want to use this. So like if you are using cloth code, probably this is easier for you to use. And I really want to just... [1:24:50] see what it can do, how we can use it in our daily lives. I would love iOS. [1:24:56] integration, like scheduling things, pushing things to my phone, like me chatting with it. And also one thing that looks like it's not super present now is storage. So currently you can store or connect it to your local computer, but...

[1:25:14] What if you're on your phone? Is there some persistence? Like there are projects in the chat side, but there are no projects in the the co-work side. So like I'm very curious how to use this, but what I love is it hooks into the skills and things I already have. So what I'm going to do is just whenever I would have started up a chat window, I'm now going to start up a co-work window [1:25:39] And just see how it behaves and goes and just, yeah, go from there. [1:25:43] Yeah, I think that's a good I think that's a good one. Okay, so we're gonna have to write the vibe check in a few and probably also do some other work. [1:25:52] I am curious if you had to give this a rating. So when we do vibe checks, we have red, yellow, green, and then we have a gold medal for paradigm shifting. So when we when we did Opus four, five, I think you and I were both at the paradigm shift level. For comparison, I think, um, [1:26:09] For GBT five, I was a [1:26:13] I was a yellow, I was a yellow green and you were a yellow, I think when it first launched, right? [1:26:18] So where do you put this if you had to give it a... [1:26:20] red yellow green or a gold medal i i would rate it from just playing with it [1:26:27] like the UI and like execution, I would say yellow because it's kind of janky. But it's very interesting. Like we can say what we want, but like it is kind of janky. But that's what they said. We made this to try out. And like I see Anthropik is very, very good at listening to people. I'm sure like someone on the team here saw me click something or you do something,

[1:26:57] But from a... [1:26:59] an idea, I would say, green. Because I really think [1:27:03] we should experiment more with the interface, what it is, and like giving this cloud code moment to more people. Like having more people that do normal work also starts to feel the paradigm shift of like async work and really handing something to an agent. And I think this could be that. Because I don't see any other company [1:27:26] do it like this. And obviously Claude is very [1:27:32] Yeah, it's a very good harness. So let's milk it. Let's make it better. But I would say ID green... [1:27:40] Execution today, right now, [1:27:44] Yellow [1:27:45] yeah i i think that's i think that's spot on and but like to to your point anthony uh morris just said i have a pr up already from something dan said so yes lol exactly yeah so that's why we do [1:27:59] yeah uh welcome to the anthropic product meeting happening live on x uh with your friendly neighborhood uh product testers um and yeah i think that i think that's that's totally right they're moving super fast i love how i love that there's already going to be changes in the product [1:28:15] For anyone who's watching or listening, you should give more feedback. Do it in the chat. [1:28:19] send it to people on X, they really actually do iterate pretty quick and

[1:28:27] I think it's fine that this is a yellow. [1:28:29] um [1:28:30] And the way that I think about it is [1:28:34] Obviously, there's a cloud code moment on X over the last couple weeks. And that was the thing that they immediately just shipped something. [1:28:41] like how many how many founders were like building a cloud code for um we were calling it cloud code easy mode because it was an ideal we were we were batting around um internally at every two and uh and they just built it they just went ahead and built it and i think that's that's so cool and it shows they really have their finger on the pulse of what people want that's that's the thing that i think is good about a lot of anthropic stuff as you can tell they use it themselves um yeah and they're using it in a very ai native and agent native way [1:29:09] Thank you. [1:29:09] Absolutely, yeah. They get it. But they're also listening. Like, they get it, but they also leave out things because they want to... [1:29:17] listen as well. They're not like, this is the way. They're like, [1:29:20] This is a way, and we're listening very carefully, and then they make iterations, which is great. [1:29:27] Yeah. Speaking of which, this is a way that if you want to build an AI, this is a way to think about doing it. We have a guide up on every about agent native architectures. So if you're a developer, this is going to be really good for a developer. It's also totally good if you're a non-developer. You just click read with Claude or read with ChatGPT and it will tell you about what this is.

[1:29:57] Graphic builds cloud code and now cloud for cloud co-work and turns it into a really easy guide to think about how do you build software in this era where agents are at the core of software instead of software being this sort of like deterministic thing that is built with deterministic code that are rules that are laid out beforehand by a programmer instead [1:30:17] The core of something like Cloud Code or Cloud Cowork is just an agent and features are really just prompts to that agent to get work done. And that opens up a whole new territory of software to build and it opens up who gets to build software. So if you're a non-technical person and you're feeling like, oh, I can't really build software, I promise you, you can. You should really try. [1:30:40] Cloud code, you should really try just a Cloud app or now. [1:30:44] It's worth trying. Cloud Cowork, I think that it could be really good for vibe coding stuff. If you're thinking about how to structure your apps, it's actually not intuitive because it's a whole new world for the way that you do programming. A lot of programmer intuition, I think, is outdated and therefore, a lot of AI intuition about how to build software is outdated. You need guides like these to help push your AI to do the right thing and build the thing in the right way. [1:31:10] I think we're getting close to time. I'm going to need to hop off. We're both going to need to hop off and get some actual work done. This is fantastic. It's so fun. We do vibe checks for every new thing that come out. We're doing these live streams as a new thing. Usually, we just write them. But you should expect next time there's a new model drop or a new product release that we will have a live stream vibe check with our internal testing. Remember, we get all this stuff before it comes out. And we'll tell you what we like and what we don't like.

[1:31:40] before we head off the stream. [1:31:42] No, cheers. Thank you, everyone. Cheers. [1:31:46] Thank you all. Check it out. Check out every. [1:31:50] Try a clock. See you. See ya. [1:32:20] insights, and laughter that will leave you on the edge of your seat. [1:32:24] craving for more. It's not just a show, it's a journey into the future with Dan Shipper as the captain of the spaceship. So do yourself a favor, hit like, smash subscribe and strap in for the ride of your life. [1:32:37] And now, without any further ado, let me just say, Dan, I'm absolutely hopelessly in love with you.