On pricing: honestly, the paid tier is less about monetization right now and more about signal. People who pay are the ones who really want this to work, and we want to build for them first. It helps us focus.
> On pricing: honestly, the paid tier is less about monetization right now and more about signal. People who pay are the ones who really want this to work, and we want to build for them first. It helps us focus.
Honestly I am pretty frugal usually but respect. I think I read something like this either in some VC idea where people will intentionally raise prices in the start to see if their market is fit as you say (or perhaps I am butchering it, but you put it out wonderfully)
I just hope that in the end, you guys become competitive to things like exe.dev which is what I currently use (for completely free mind you but its in alpha version) but it provides an agent shelly with the web in a sandbox and I think that they are gonna focus on being more competitive in the market and there are multiple of options right now
Though just now I saw that 1code.dev is an YC backed startup and um, being honest, I have just less faith in YC backed/VC funded startups unless proven otherwise because VC requires profits so they might be incentivized to keep the pricing on the upper side of things and not change as you mention
I am not exactly sure how things would go for your product and I am unable to get a clear visual understanding of what you are trying to achieve but I feel like there are some things which you can do to solve it as the GP said
Basically please dont get enshittified in the end.
With Claude Code now available in the Claude desktop app, what are the key differentiators?
I’m all for new, better tools and will probably test this out over the weekend to see if the interface is more usable than Claude’s. Just curious if there’s anything you’re particularly excited about compared to the official release.
We started building this after trying existing tools like Claude Code desktop. We realized we couldn't fully transition to them — my co-founder went back to Cursor and burned tons of money on tokens.
So we decided to build a tool that WE would actually use, and once we succeeded with that, we open-sourced it.
Since then we've received a lot of feedback on where to take this product - some use cases Anthropic will probably never do, like integrating Codex or adding other z.ai GLM subscription.
We're just trying to ship something we use ourselves, and maybe it could be useful to others, and see where it goes.
fyi: it does not build for me from the source code.
Rebuild Failed
An unhandled error occurred inside electron-rebuild
node-gyp failed to rebuild '[...]/1code/node_modules/node-pty'
Error: node-gyp failed to rebuild '[...]/1code/node_modules/node-pty'
at ChildProcess.<anonymous> ([...]/1code/node_modules/@electron/rebuild/lib/module-type/node-gyp/node-gyp.js:121:24)
at ChildProcess.emit (node:events:508:28)
at ChildProcess._handle.onexit (node:internal/child_process:294:12)
node:child_process:1000
throw err;
^
Could you fix your video to not zoom in and out constantly like that? It's pretty hard to tell what's going on when it zooms right in every time you click something, and it's so fast it actually gave me a bit of motion sickness.
I’m happy to see another project monetizing built executables. I feel like it’s not a popular option to many, but to me I think it’s the best way to getting financial support from enterprises that would be more than happy to leech.
Thanks! for us monetization is more about signal - people who spend $20 are the ones who really want this to work, and we want to build for them. We have a private Discord with our Pro users where we discuss what to build next, share feedback and tips and tricks.
What's the story on local-only operation with a paid subscription? If it works like it seems to from the screen shots and video, I'm more than happy to pay for this even while using the OSS version (and I probably am going to modify it myself quite a bit for my own use-cases), but in my work environment:
1. Absolutely nothing can be sent off-site
2. All AI API requests must go through our custom gateway (we've got deals with all the major AI providers, and I think that even involves a degree of isolated hosting in specific approved cloud environments)
While $20/mo feels a bit weird for an app that doesn't rely on its own cloud service, I'd subscribe right now just because I do "really want this to work".
i would 100% pay for non-electron/non-vscode ui (sublime text? anyone?)
nothing specifically against vscode/electron/atom. i did use atom 10 years ago, then started sublime text 2, since then, it has been really hard to beat sublime text. the speed and rendering of sublime is much better.
i also have been using zed, which can be considered cloud-code ui too in certain terms, zed is quite more resource-intensive than the sublime and feels higher latency. (possibly because I use frosty-transparent theme...)
What is your use case that you see UI lag between vscode and sublime? Honestly, I feel zero difference between sublime/vscode/vi. Vscode arguably takes longer to boot up, but that only happens like once a day so it's not a big deal.
I think this is a lot of "I don't like Typescript/Javascript for serious things" or "Electron sucks" posturing rather than an actual tangible difference.
If you don't feel these differences every keystroke, count yourself lucky to have slower perception or typing, rather than accusing folks of posturing.
Your brain processes (visual) information at a resolution of >= 80ms[1]. The idea that you can tell the difference between 10ms or 50ms latency when typing is simply untrue (both events will appear instantaneous). I say this as someone that has played Counter-Strike professionally and have a sub-200ms reaction time. (Auditory perception is processed at a higher resolution, but the article is decidedly not about that.)
i cannot tell exactly but it kinda bothers me while working/typing
it is not like a huge latency, definitely not like ssh-connection.
to explain better, i usually have pre-defined set of keystrokes i input, so it's not the issue of latency of a single keystroke, rather compounding effect.
another thing is, most of the LSPs, highlighting etc are visibly slower on vscode. I am also having many plugins/extensions so that is partly to blame.
in the recent versions of vscode, they started supporting tree-sitter, which is quite nice in terms of performance.
Zed is awesome and zed has zed AI and its built on non electron. Its actually built by the atom team itself by learning from their mistakes of building atom and this time instead of building it on browser, they did some pretty hardcore optimizations like running the editor at 144 fps or similar by rendering rectangles via GPU processing.
Zed is really snappy. It's kind of my go-to editor for most things.
The web version runs Claude Codes in remote sandboxes - we're adding the same to macOS soon.
Beyond that, we're working on API & integrations - connecting Sentry so an agent can automatically open a PR to fix an issue, or triggering Claude Codes directly from @ mentioning in GitHub/Slack/Linear.
Hmm. But the value of using claude code for me is that it runs on my local machine. I have setup several skills / commands / workflows that rely on the content on my machine. So i guess using it on a remote sandbox won't help me.
>Yes, 1Code works with your Claude Pro or Max subscription. Sign in with your Anthropic account and use your existing subscription. We charge separately for the 1Code app.
Is this even after the recent events; Anthropic banning 3rd-party access to the Claude Pro/Max subscriptions and only allowing API access? Tool is nice but most certainly not* 800$ nice.
*Based on heavy Claude Max users here, the people that will be running 3-4 agents, saying API would've cost them $1k+.
Can you expand on this further? You’re using a sandbox to “spoof” authentication by intercepting the auth token to a web browser? You use the agent-sdk which is fine but youre supposed to ONLY use that via api I thought?
Ah, I see. My only other caveat (other models/agents) is addressed in upcoming work to do. Otherwise, looks good, simple/straightforward UI, functional. Great work overall.
Okay, so I want to like it and I think there is potential here.
I may be missing something in my setup, but I find these things happening in all Claude wrappers I have tried.
1. After a long stint of planning or back and forth exchange, I have to scroll to find the bottom of the conversation. Same with Claude itself.
2. I don’t want tabs as much as I want split window’s that I can label and highlight.
3. The more calm it is, I feel like I’m going to miss something that it’s fucking up. Sometimes in normal CC, I will look over and it’s making a large assumption and nuking a change we just made. I would love to control the verbosity or at least be able to peak behind the curtain at certain points.
4. I like the click flow for the planning Q&A but it wasn’t clear if I could add my own answer like I can in CC.
Other than these, I like it and want to use it more so than my current daily driver CLIManager. (Which has the overlap with points 1 and 2)
Thank you for the detailed feedback! Really appreciate you taking the time to share these points.
On point 4 (custom answers in Q&A): You're right, this is missing! I'll add the ability to write your own answers today or tomorrow.
On point 1 (scrolling): I pushed a batch of scroll fixes yesterday, but I see it's still not working optimally. We're actively working on improving this - it's definitely on our radar.
On point 2 (split windows): We're planning to add multi-window support soon - so you can have multiple chat windows open side by side, similar to how you'd work with multiple terminal windows, but with a nice UI. Stay tuned!
On point 3 (verbosity control): Great suggestion - sometimes you want to catch assumptions before they cause damage. Adding to our roadmap.
Thanks again, and glad you're finding it useful despite these rough edges!
I hate to be that guy, but if anyone got excited about an interface where they could work with Claude, I'd say stick with the cli.
If you really want a 'different' UI thats unlike VSCode or Cursor already, I'd genuinely recommend checking out almost any other tool that pairs with CC.
Are there docs somewhere? I was looking to see if forgejo/gitlab/bitbucket are supported or if it's only github? Or if it even opens PRs on its own or not
The pricing is ridiculous. It doesn't include the Claude subscription so $20/m is way out of league for a UI.
And a "max" subscription for $200/m??? It makes no sense. It's not because LLM making companies have plans at $200 that any company should make a $200/m plan.
"Claude Code, but usable" might not be the best tagline. I find Claude Code extremely usable. In fact, it's the simplest product that "just works" I've ever used.
The tagline might feel off if you're already comfortable with CLI. Though there's definitely an audience for this — my co-founder posted about struggling with CLI and it got 240k views with 400+ replies of people feeling the same: https://x.com/serafimcloud/status/2002304990053085542
Even as a technical person, I remember my transition from Cursor to Claude Code CLI being rough. Maybe 1Code could be a way to get the best of both worlds, power of claude code and UX of Cursor
noted, title comes from my cofounder who's not technical, but he loved Cursor, I was convincing Serafim to try Claude Code, but he just couldn't use it. Cause it was a CLI, then I created a first alpha version of this UI for them .... and he loved it! he's now a fan of Claude Code, and that's where reasoning for "useful" coming from
The idea is nice, but the pricing is odd. 20$/month (the cost of Claude Pro) just for a web interface and a sandbox seems very expensive.
I hope the business model works out for you, but I doubt the price is justified.
Also a better link is: https://1code.dev/
The repo doesn't have screenshots, so you might loose traffic just because of that.
Nevertheless, I'll try it out - looks nice!
Thanks for the feedback!
On pricing: honestly, the paid tier is less about monetization right now and more about signal. People who pay are the ones who really want this to work, and we want to build for them first. It helps us focus.
> Thanks for the feedback!
> On pricing: honestly, the paid tier is less about monetization right now and more about signal. People who pay are the ones who really want this to work, and we want to build for them first. It helps us focus.
Honestly I am pretty frugal usually but respect. I think I read something like this either in some VC idea where people will intentionally raise prices in the start to see if their market is fit as you say (or perhaps I am butchering it, but you put it out wonderfully)
I just hope that in the end, you guys become competitive to things like exe.dev which is what I currently use (for completely free mind you but its in alpha version) but it provides an agent shelly with the web in a sandbox and I think that they are gonna focus on being more competitive in the market and there are multiple of options right now
Though just now I saw that 1code.dev is an YC backed startup and um, being honest, I have just less faith in YC backed/VC funded startups unless proven otherwise because VC requires profits so they might be incentivized to keep the pricing on the upper side of things and not change as you mention
I am not exactly sure how things would go for your product and I am unable to get a clear visual understanding of what you are trying to achieve but I feel like there are some things which you can do to solve it as the GP said
Basically please dont get enshittified in the end.
Thanks for the feedback. We’ve added a short demo video to the repo to clarify the UX and direction. Our focus is earning trust by shipping.
With Claude Code now available in the Claude desktop app, what are the key differentiators?
I’m all for new, better tools and will probably test this out over the weekend to see if the interface is more usable than Claude’s. Just curious if there’s anything you’re particularly excited about compared to the official release.
We started building this after trying existing tools like Claude Code desktop. We realized we couldn't fully transition to them — my co-founder went back to Cursor and burned tons of money on tokens.
So we decided to build a tool that WE would actually use, and once we succeeded with that, we open-sourced it. Since then we've received a lot of feedback on where to take this product - some use cases Anthropic will probably never do, like integrating Codex or adding other z.ai GLM subscription.
We're just trying to ship something we use ourselves, and maybe it could be useful to others, and see where it goes.
Why not support open code as well?
its in the backlog
fyi: it does not build for me from the source code.
Rebuild Failed
An unhandled error occurred inside electron-rebuild node-gyp failed to rebuild '[...]/1code/node_modules/node-pty'
Error: node-gyp failed to rebuild '[...]/1code/node_modules/node-pty' at ChildProcess.<anonymous> ([...]/1code/node_modules/@electron/rebuild/lib/module-type/node-gyp/node-gyp.js:121:24) at ChildProcess.emit (node:events:508:28) at ChildProcess._handle.onexit (node:internal/child_process:294:12) node:child_process:1000 throw err; ^
- you should replace monaco here https://github.com/21st-dev/1code/blob/47b72fb88bd15cf01cb91... with codemirror 6
- not only is it much more performant but also mobile friendly
thanks
Could you fix your video to not zoom in and out constantly like that? It's pretty hard to tell what's going on when it zooms right in every time you click something, and it's so fast it actually gave me a bit of motion sickness.
i noticed all these new ai-tool launches have this zooming effect. maybe 1.2x zoom is fine, but these usually over-do the zoom like 3x.
instead, an old-school arrow/pointer (or pointer trails like its '98!) would be nicer.
I’m happy to see another project monetizing built executables. I feel like it’s not a popular option to many, but to me I think it’s the best way to getting financial support from enterprises that would be more than happy to leech.
Thanks! for us monetization is more about signal - people who spend $20 are the ones who really want this to work, and we want to build for them. We have a private Discord with our Pro users where we discuss what to build next, share feedback and tips and tricks.
What's the story on local-only operation with a paid subscription? If it works like it seems to from the screen shots and video, I'm more than happy to pay for this even while using the OSS version (and I probably am going to modify it myself quite a bit for my own use-cases), but in my work environment:
1. Absolutely nothing can be sent off-site
2. All AI API requests must go through our custom gateway (we've got deals with all the major AI providers, and I think that even involves a degree of isolated hosting in specific approved cloud environments)
While $20/mo feels a bit weird for an app that doesn't rely on its own cloud service, I'd subscribe right now just because I do "really want this to work".
i would 100% pay for non-electron/non-vscode ui (sublime text? anyone?)
nothing specifically against vscode/electron/atom. i did use atom 10 years ago, then started sublime text 2, since then, it has been really hard to beat sublime text. the speed and rendering of sublime is much better.
i also have been using zed, which can be considered cloud-code ui too in certain terms, zed is quite more resource-intensive than the sublime and feels higher latency. (possibly because I use frosty-transparent theme...)
What is your use case that you see UI lag between vscode and sublime? Honestly, I feel zero difference between sublime/vscode/vi. Vscode arguably takes longer to boot up, but that only happens like once a day so it's not a big deal.
I think this is a lot of "I don't like Typescript/Javascript for serious things" or "Electron sucks" posturing rather than an actual tangible difference.
> What is your use case that you see UI lag between vscode and sublime? … I think this is … posturing …
Typing with pleasure: https://pavelfatin.com/typing-with-pleasure/
Study the graphs. Ready the copy.
If you don't feel these differences every keystroke, count yourself lucky to have slower perception or typing, rather than accusing folks of posturing.
Your brain processes (visual) information at a resolution of >= 80ms[1]. The idea that you can tell the difference between 10ms or 50ms latency when typing is simply untrue (both events will appear instantaneous). I say this as someone that has played Counter-Strike professionally and have a sub-200ms reaction time. (Auditory perception is processed at a higher resolution, but the article is decidedly not about that.)
[1] https://pmc.ncbi.nlm.nih.gov/articles/PMC9851611/
i cannot tell exactly but it kinda bothers me while working/typing
it is not like a huge latency, definitely not like ssh-connection.
to explain better, i usually have pre-defined set of keystrokes i input, so it's not the issue of latency of a single keystroke, rather compounding effect.
another thing is, most of the LSPs, highlighting etc are visibly slower on vscode. I am also having many plugins/extensions so that is partly to blame.
in the recent versions of vscode, they started supporting tree-sitter, which is quite nice in terms of performance.
I almost noted on that as I've harped on sub-200ms for web responsiveness since the 90s.
However, reacting to something you see is a diff thing than sensing intra key-to-char lag in flow.
In CS, responding to what's on screen is diff from button press to seeing game action. High polling rate controls are going after that.
Zed is awesome and zed has zed AI and its built on non electron. Its actually built by the atom team itself by learning from their mistakes of building atom and this time instead of building it on browser, they did some pretty hardcore optimizations like running the editor at 144 fps or similar by rendering rectangles via GPU processing.
Zed is really snappy. It's kind of my go-to editor for most things.
I'm already using it I heard it from X
It's cool first time I see it I say NO WAY!!!
LOVE IT SO MUCH
I didn't quite understand what "Unlimited web & macOS apps access" means in the paid tier. Could you elaborate?
The web version runs Claude Codes in remote sandboxes - we're adding the same to macOS soon.
Beyond that, we're working on API & integrations - connecting Sentry so an agent can automatically open a PR to fix an issue, or triggering Claude Codes directly from @ mentioning in GitHub/Slack/Linear.
Hmm. But the value of using claude code for me is that it runs on my local machine. I have setup several skills / commands / workflows that rely on the content on my machine. So i guess using it on a remote sandbox won't help me.
>Do I need a Claude subscription?
>Yes, 1Code works with your Claude Pro or Max subscription. Sign in with your Anthropic account and use your existing subscription. We charge separately for the 1Code app.
Is this even after the recent events; Anthropic banning 3rd-party access to the Claude Pro/Max subscriptions and only allowing API access? Tool is nice but most certainly not* 800$ nice.
*Based on heavy Claude Max users here, the people that will be running 3-4 agents, saying API would've cost them $1k+.
1Code uses the official Claude Code SDK (https://platform.claude.com/docs/en/agent-sdk/overview) which wraps the Claude Code binary. This is the official way to build on top of Claude Code.
Some other tools (like OpenCode) made direct API calls while impersonating Claude Code - that approach got banned. We're not doing that.
Can you expand on this further? You’re using a sandbox to “spoof” authentication by intercepting the auth token to a web browser? You use the agent-sdk which is fine but youre supposed to ONLY use that via api I thought?
Ah, I see. My only other caveat (other models/agents) is addressed in upcoming work to do. Otherwise, looks good, simple/straightforward UI, functional. Great work overall.
Okay, so I want to like it and I think there is potential here.
I may be missing something in my setup, but I find these things happening in all Claude wrappers I have tried.
1. After a long stint of planning or back and forth exchange, I have to scroll to find the bottom of the conversation. Same with Claude itself.
2. I don’t want tabs as much as I want split window’s that I can label and highlight.
3. The more calm it is, I feel like I’m going to miss something that it’s fucking up. Sometimes in normal CC, I will look over and it’s making a large assumption and nuking a change we just made. I would love to control the verbosity or at least be able to peak behind the curtain at certain points.
4. I like the click flow for the planning Q&A but it wasn’t clear if I could add my own answer like I can in CC.
Other than these, I like it and want to use it more so than my current daily driver CLIManager. (Which has the overlap with points 1 and 2)
Nice work overall!
Thank you for the detailed feedback! Really appreciate you taking the time to share these points.
On point 4 (custom answers in Q&A): You're right, this is missing! I'll add the ability to write your own answers today or tomorrow.
On point 1 (scrolling): I pushed a batch of scroll fixes yesterday, but I see it's still not working optimally. We're actively working on improving this - it's definitely on our radar.
On point 2 (split windows): We're planning to add multi-window support soon - so you can have multiple chat windows open side by side, similar to how you'd work with multiple terminal windows, but with a nice UI. Stay tuned!
On point 3 (verbosity control): Great suggestion - sometimes you want to catch assumptions before they cause damage. Adding to our roadmap.
Thanks again, and glad you're finding it useful despite these rough edges!
Thank you for the meaningful reply. I’ll be watching the repo for updates
Have you tried https://air.dev/ ?
Have not, but interesting offering from JetBrains. I’ll poke at it with a stick. Thanks for the suggestion
You shouldn’t start filenames or identifiers with digits.
why
I hate to be that guy, but if anyone got excited about an interface where they could work with Claude, I'd say stick with the cli.
If you really want a 'different' UI thats unlike VSCode or Cursor already, I'd genuinely recommend checking out almost any other tool that pairs with CC.
Are there docs somewhere? I was looking to see if forgejo/gitlab/bitbucket are supported or if it's only github? Or if it even opens PRs on its own or not
The pricing is ridiculous. It doesn't include the Claude subscription so $20/m is way out of league for a UI.
And a "max" subscription for $200/m??? It makes no sense. It's not because LLM making companies have plans at $200 that any company should make a $200/m plan.
"Claude Code, but usable" might not be the best tagline. I find Claude Code extremely usable. In fact, it's the simplest product that "just works" I've ever used.
Yeah, this is a weird angle to go down when you're trying to compete against... the most popular product.
The tagline might feel off if you're already comfortable with CLI. Though there's definitely an audience for this — my co-founder posted about struggling with CLI and it got 240k views with 400+ replies of people feeling the same: https://x.com/serafimcloud/status/2002304990053085542
Even as a technical person, I remember my transition from Cursor to Claude Code CLI being rough. Maybe 1Code could be a way to get the best of both worlds, power of claude code and UX of Cursor
I feel like you're trying to straddle two worlds, like Jean-Claude van Damme. Easier to pick a lane! Good luck.
noted, title comes from my cofounder who's not technical, but he loved Cursor, I was convincing Serafim to try Claude Code, but he just couldn't use it. Cause it was a CLI, then I created a first alpha version of this UI for them .... and he loved it! he's now a fan of Claude Code, and that's where reasoning for "useful" coming from
> Mom I want 1Code
> Stop crying son, we have 1Code at home
> Shows tmux
lol