1Code - Open source Cursor-like UI for Claude Code
by•
Whats 1Code?
An app to run your Claude Code agents in parallel that works on Mac and Web.
On Mac - run locally, with or without worktrees.
On Web - run in remote sandboxes with live previews of your app, mobile included, so you can check on agents from anywhere.
Running multiple Claude Codes in parallel dramatically sped up how we build features.
Replies
21st.dev
@serafimcloud congrats on the launch! Does it work with the regular Anthropic subscription as well, or only Claude Code?
21st.dev
@serafimcloud @austin_heaton since we're utilizing Claude Code CLI under the hood, we accept everything that Claude Code CLI accepts, and as I know they only work either with API key or with Claude Code subscription
PicWish
@serafimcloud Such a necessary step! Are you keeping it as a UI layer or planning to add your own logic on top?
@serafimcloud for this. How do you manage around the 5 hour and weekly limits? For us 2 devs, in 2 projects, it gets exhausted all the time. So that's a challenge I don't have an answer to!
@serafimcloud Parallel agents only matter once you can trust them.
GTWY is built around the same idea: scale intelligence by making execution predictable, not fragile.
OpenCommit: git commit message by AI
finally.. the TUI always jumps and jiggles, idk why anthropic cant fix it
21st.dev
@sukharev made it for you!
Community Figma MCP server
Will Anthropic ban it similarly to Open Code once it becomes popular?
21st.dev
@anton_tishchenko 1Code uses the official Claude Code SDK 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.
NO WAY!!!
LOVE IT SO MUCH
21st.dev
@sayyidalijufri thanks!
Huge congrats on the launch! Love the focused, parallel Claude Code workflow for serious shipping.
21st.dev
@zeiki_yu thanks!
ProblemHunt
Wow!
what a launch!!!
21st.dev
@a_6 🚀🚀🚀
Niiiiice @serafimcloud @serjobas it got even better than the alpha versions that I've tested. Waiting for browser preview so I don't have to jump between agents and my browser anymore (:
No Cap
Amazing, much needed
This is a solid and timely contribution to the AI coding space. Running multiple Claude Code agents in parallel with both local and remote sandbox support is a strong productivity unlock, especially with live previews across devices. The open source approach also signals long term intent and trust. Insight wise, parallel agents change not just speed but how developers reason about progress and ownership across tasks. From a technical and onboarding perspective, how do you coordinate state, conflicts, and context sharing between parallel agents, and what signals do you surface to help new users understand which agent is doing what during their first sessions?