正在加载视频...

视频加载失败

Just open sourced Monocode - a desktop UI for your coding agents. Works with your existing Claude Code, Codex, Cursor, and OpenCode harnesses. Tabs can be sessions, a lightweight editor or a terminal. Currently only supports macOS Apple Silicon. Still very early so expect bugs. Feedback and small PRs...

53,624 次观看 • 1 个月前 •via X (Twitter)

70 条评论

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

link to the source

zifu 的头像
zifu1 个月前

When will Pi support come?

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

Working on it :)

zifu 的头像
zifu1 个月前

Thanks! Also, will you ever release your agent macos status bar project as well?

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

Ahh yes 😁 need to clean it up a bit

Apollon Foivos Bakis 的头像
Apollon Foivos Bakis1 个月前

Having one UI across different coding agents is really useful.

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

certainly and its not a new idea... many alternatives out there :)

Jakub Łapeta 的头像
Jakub Łapeta18 天前

Are you thinking about mobile app? It would be great to see one! When it comes to bugs: sometimes if I switch models inside one session the second one can’t load the context of the conversation. It happened when I used Fable for planning, few tasks and later switched to sonnet

Nikolay Petkov 的头像
Nikolay Petkov18 天前

Yeah I just discovered it as well it’s a bug and it’s fixed in The latest commit… I will make a release in a couple of hours s

Jakub Łapeta 的头像
Jakub Łapeta18 天前

What about that mobile app? Are you planning one?

Nikolay Petkov 的头像
Nikolay Petkov18 天前

yeah, definitely. i want to get a few other things right first before the mobile app

丂ムんノレ 的头像
丂ムんノレ1 个月前

linux when?

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

Soon :)

丂ムんノレ 的头像
丂ムんノレ1 个月前

umm i made the change and it work in linux

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

thats awesome, you can open a pr if you want but please check CONTRIBUTING.md before you do fyi I am the only maintainer :)

丂ムんノレ 的头像
丂ムんノレ1 个月前

sure sure i will do build check

Ben Mo 的头像
Ben Mo14 天前

man this is brilliant!

Nikolay Petkov 的头像
Nikolay Petkov14 天前

thanks 🙏

Coach Mov 的头像
Coach Mov25 天前

context 창이 계속 그대로야. 새로운 세션을 만들어도 초기화가 안되는 현상이 있어.

Sandeep 的头像
Sandeep17 天前

when can i expect a windows version or is it already out there? sorry the github instructions were not so clear hence clarifying

Nikolay Petkov 的头像
Nikolay Petkov17 天前

Yeah windows is out check the releases page

Sandeep 的头像
Sandeep17 天前

thank you. one more clarification: is there any telemetry or logs that you are centrally logging? just checking from an enterprise usage standpoint.

Nikolay Petkov 的头像
Nikolay Petkov17 天前

No none :)

Basil Khowaja 的头像
Basil Khowaja1 个月前

Does it support file manager inside the app so I can view almost all files in a specific project path that i open up?

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

Yeah it should also auto update if the agent is currently adding or deleting files too

Basil Khowaja 的头像
Basil Khowaja1 个月前

Ahh, I see. When do you intend to release linux supportV

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

there are still bugs i want to iron out but linux support is coming :)

Buswe 的头像
Buswe1 个月前

Tabs as isolated sessions is the sneaky-good part, two agents can't share state and step on each other while you keep both in view.

probe 的头像
probe1 个月前

Tabs as sessions is the part I want most. Jumping between a terminal and an agent run in one window is where my setup gets messy, so this fits nicely.

Francis 的头像
Francis23 天前

Incredible UI and the performance seems great so far. Curious how you are thinking about workspaces. So far I cannot see a way to create them within the app itself.

pdp 的头像
pdp1 个月前

It is beautiful.

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

Thank you :)

Paresh Dudhat 的头像
Paresh Dudhat1 个月前

Looks pretty clean ui. Also building something similar. Just want to know why you chose Tauri?

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

mostly for size and speed... monocode is 25mb while similar electron apps are 500mb performance so far is pretty great as well but you could probably achieve the same with electron... idk

elias 的头像
elias11 天前

Super cool. Mobile app?

Nikolay Petkov 的头像
Nikolay Petkov11 天前

on the roadmap... in the next couple of weeks i would say :)

Andika 的头像
Andika28 天前

This is great! Will u someday make it like agentic chat UI similar like grokbot? That'll be interesting

Nikolay Petkov 的头像
Nikolay Petkov28 天前

that would be probably a project on its own

Andika 的头像
Andika28 天前

interesting, can’t to see that future project started! keep up the good work🙌

uhhexe 的头像
uhhexe22 天前

What exactly is this? Looks insanely useful. Is this a coding environment? Looks great

Nikolay Petkov 的头像
Nikolay Petkov22 天前

its a GUI for your coding agents... if you already use codex, CC cursor and many others you can run them in Monocode so you dont have to switch apps or terminals not a new idea i just built it cause i wanted something for myself

Thomas Nguyen 的头像
Thomas Nguyen1 个月前

So it like cmux, conductor, can we orchestrate, spawn lanes?

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

not currently but definitely sth i am thinking about

Dan 的头像
Dan28 天前

Is it like @orca_build but early stage?

Nikolay Petkov 的头像
Nikolay Petkov28 天前

@orca_build it depends… if the main idea is to work with multiple harnesses from one interface then yes

Lily|AI×自動化×マーケティング🤖 的头像
Lily|AI×自動化×マーケティング🤖1 个月前

タブがセッションにもエディタにもなるの便利そうですね。僕もClaude Code複数走らせて見失うので試します

Felipe Orlando 的头像
Felipe Orlando1 个月前

looks amazing... two things that any other app miss for me are: - orchestration, like Orca has - spaces (personal, work, oss, etc), like... Arc?

安叫兽|Bird🕊️ 🔶 BNB 的头像
安叫兽|Bird🕊️ 🔶 BNB1 个月前

标签页能混开会话和终端,这个思路挺顺手。

Rayel 的头像
Rayel1 个月前

Are you planning to support grok?

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

yeah grok is next! funny enough it was in the initial version of the product but i scrapped it

a 的头像
a28 天前

When windows support?

Nikolay Petkov 的头像
Nikolay Petkov28 天前

will review today

Wim 的头像
Wim1 个月前

What makes it different from e.g t3code? Any plans to support pi or omp or GitHub copilot?

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

i would say the same like t3code and in all honesty t3code is probably more featureful pi coming next... haven't thought about omp or copilot

Wim 的头像
Wim1 个月前

Would recommend. Everyone else is already doing Opencode, Codex, Claude while a lot use pi or omp. Copilot is great too, although people bash it because it’s GitHub but I probably like it the most. Just my 2 cents

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

i agree. also i was an early Copilot customer and it was pretty good but their price hike made me switch over

Wim 的头像
Wim1 个月前

Have a look at copilot SDK

Ihor Kolobanov 的头像
Ihor Kolobanov25 天前

Nice, congrats! I'm building - similar idea, but built around multi-harness workflows: Claude Code plans and delegates, Codex implements, opencode reviews. Are tabs staying independent by design, or is cross-session coordination on the roadmap?

Nikolay Petkov 的头像
Nikolay Petkov25 天前

thanks warpforge looks cool not sure yet but orchestration is definitely on the roadmap

Ihor Kolobanov 的头像
Ihor Kolobanov25 天前

Cool, will be watching. Happy to compare notes when you get to it

Barth 的头像
Barth1 个月前

Looks pretty nice! I'm desperately looking for the perfect tool to orchestrate all my agents from Cursor, Codex, Claude and OhMyPi. Tried many ADE like Superset, Orca, Conductor, Paseo, T3 Code etc. but only herdr can do what I want for now, but I want something nicer to use.

Shreyansh | Building AI Systems 的头像
Shreyansh | Building AI Systems1 个月前

A shared interface for coding agents could reduce tool switching, but the durable value is a consistent review surface. Plans, diffs, commands, tests, and approvals should look familiar even when the underlying model changes.

AI Mastery Guide 的头像
AI Mastery Guide1 个月前

That UI looks super clean

maddada • ghostex.dev 的头像
maddada • ghostex.dev25 天前

Looks great man, very polished and smooth ux. I'm like a year into the ADE journey and my honest advice is it's not worth it 😂 There's like 1 billion of them and it's impossible to make money working on yet another ADE. Also the adrressable market is somewhat small. If you really want to keep implementing then send me a dm I can share stats about what people actually use so you focus on those things :) All the best.

Jul.io 🇩🇪 的头像
Jul.io 🇩🇪20 天前

Man it looks great! What about the performance, RAM, CPU Connectivity using SSH?

Thomas 的头像
Thomas1 个月前

yet another mac only app

Nikolay Petkov 的头像
Nikolay Petkov1 个月前

sowwy 🥺 linux and windows coming soon

LocalAIMaxxer // Niko 的头像
LocalAIMaxxer // Niko24 天前

My Bulgarian brother, surely you can't have the landing page like this in the age of AI. I want to see more about Monocode

Nikolay Petkov 的头像
Nikolay Petkov24 天前

in the age of AI you don’t need a landing page 😊 Jk I will do it at some point

LocalAIMaxxer // Niko 的头像
LocalAIMaxxer // Niko24 天前

Add it! And let's link up at some point, I'm also Nikolay, in London, working in AI.

相关视频

Another insane Jev use case! Jev is making it dramatically cheaper to evaluate what actually happened inside an agent run. And finally, someone open-sourced a self-improving memory layer that can put that signal to work across agent harnesses: - Claude Code - Codex - Cursor - OpenCode, and 20+ more Beacon by Asymptote Labs continuously captures your agent history across harnesses and uses Jev to identify which runs are actually worth learning from. It then turns the highest-signal workflows, corrections, and debugging patterns into reusable skills. GitHub repo: (don’t forget to star it ⭐ ) Beacon preserves the complete session history. But preserving a run and learning from it are two different things. Most coding-agent sessions contain routine exploration, failed commands, and fixes that only apply to one task. The trace can remain available for inspection without turning every detail into guidance for future agents. Jev scores each run for evidence, reuse potential, and human correction signals. An application policy then decides whether to promote, review, or discard it. The recording shows this in action. Claude receives a coding task, modifies the implementation, and runs the tests. I then provide an edge-case correction, so Claude updates the code and adds regression coverage. Beacon automatically captures the complete session. Jev evaluates whether the correction contains a reusable engineering lesson. Once approved, that lesson becomes available to other coding agents working on the project. Since it works across harnesses: - Claude Code sessions can teach Codex. - Cursor debugging can improve OpenCode. So a problem solved by one agent should not need to be learned from scratch by another. If you want to dive deeper into Jev, I also wrote a hands-on guide to building this Jev-style decision path with open models, entirely locally. Read it below.

Avi Chawla

289,832 次观看 • 9 天前

Another insane Jev use case! Jev makes it incredibly cheap to evaluate and classify agent runs at scale. And finally, someone open-sourced a self-improving memory layer that can put that capability to work across agent harnesses. It turns your agent sessions into a compounding knowledge layer, where every successful run can make future agents smarter across: - Codex - Claude Code - Cursor - OpenCode and 20+ more Beacon by Asymptote Labs continuously builds a shared history across your agent harnesses and uses Jev to identify the runs worth learning from. It then turns the best workflows, corrections, and debugging patterns into reusable skills. GitHub repo: (don’t forget to star it ⭐) Most agent runs are messy. They contain exploration, failed commands, dead ends, and one-off fixes that should never become permanent memory. So Beacon preserves the full session history, while Jev helps decide what should be promoted, reviewed, or discarded. The recording below shows this in action. Beacon found 579 sessions across 5 coding-agent harnesses and normalized them into one consistent history. From there, Jev surfaces the lessons worth keeping and makes them available across your agent stack. - A pattern learned in Cursor can carry into OpenCode. - A lesson from Claude Code can improve the next Codex run. Every successful run adds to the shared knowledge layer, making future agents smarter. If you want to dive deeper into Jev, I also wrote a breakdown of how it works. The article is quoted below.

Akshay 🚀

116,880 次观看 • 7 天前