> A native home for Claude Code, Codex and every coding agent. Every pixel is a file your agent can change. Close it, they keep working.
First two sentences give zero information what it does. Third one is useful but i think many harnesses do it?
Continuing:
> A new kind of agent workspace. Native and terminal first, with nothing between you and your agents. Every part of it is a plain folder your agent can rewrite, and every agent keeps running when the window is gone.
Okay, more of the same. Let's press on.
> Your agents run agents.
And hear back.
> An agent hands work to helpers and keeps going. When they finish, their reports land in its input. No copy-paste between windows: herdr-core carries the messages.
Are there modern harnesses that don't do subagents/subthreads/whatever the same way?
---
There are some interesting bits but the vibed copy is for me offputting and a sign even the author might not understand what he made.
Yeah that copy is pretty bad. My understanding (someone correct me if I'm wrong), but the herdr native app is an improvement over the existing TUI multiplexer, with a new API for facilitating comms between agents. Herdr is really useful if you've not tried it. I'll be giving this a go (I hadn't really been using agent comms between panes anyway, more just for workspace navigation). Seems you can think of herdr as a 'harness for your harnesses'!
I can't speak to the GUI, just the TUI compared to tmux.
herdr is very flexible, but it shines when you are working with multiple projects, with multiple worktrees, and multiple agents within those projects.
1. herdr uses workspaces to help organize tabs. Conceptually, a workspace probably maps to a project, but it doesn't have to.
2. herdr has native worktree support, and can create a nested workspace per worktree with a keyboard or CLI command. It can check out worktrees for you
3. herdr identifies agents across tabs and workspaces and pulls them out into an extra navigational pane. It shows the status of those agents (e.g. idle, working, waiting, done), and can play a sound when they change state. This helps you track multiple agents across multiple projects and quickly switch to them when they need help without needing to remember "claude is running in space 2, tab 3"
4. herdr integrates with your agents, which can control herdr's UI and rename tabs based on their status.
5. All this is scriptable via the herdr CLI, and agents are very good at it. But it also has full mouse support, if you're new to multiplexers its very easy to start interacting with it.
I usually keep one tmux session per Python worktree, with a local Postgres test container in another pane. The status view sounds useful here; does Herdr detect waiting via agent hooks or infer it from terminal output?
> What's useful about Herdr compared to tmux and claude code or codex's built in sub-agents?
Agent status detection. Single view over local and remote agents. /herdr skill that lets agents list other agents, peek into their panes, send keys (yes, you can make a lot of it work with tmux and some elbow grease, but herdr provides nice cli/api for it and works out of the box) including to remote ones.
In my mental model, subagents are also different from top level agents in tabs: subagents are only great as temporary workers you never want to interact with.
It also lets you message between Claude Code and Codex sessions. If you're using Claude Code, their built-in messaging system only can see/message other Claude Code sessions (for obv reasons).
This is an unlock, because these companies will (prob) never build this feature.
I think i built something similar after herdr was released but i can say the reason is at least
tmux is definitely usable but moving between workspaces/repos can be annoying if youre managing mahy many things at once. and difficult to organize. more over having agents manage other agents via tmux can lead to garbled input/output
i havent used codex. claude subagents i think arent the right comparison - the closer comparison would be claude managed agents. but i imagine one reason herdr might be better would be that you’re not stuck in the anthropic ecosystem
If you prefer tmux orchestration you could give my plugin a try.
It's just a few bash scripts enabling agents to message each other. Works with the desktop apps as well so you could have a Codex or Claude desktop app orchestrate tmux sessions.
Integrated experience. You can connect to multiple machines and access terminals on them through spaces in the left nav. Agents across all spaces are listed in a dedicated section for quick access.
Each space is persistent contains persistent terminals like tmux, so disconnections do not affect them.
I'm also curious, if you just ask Codex, Claude etc already do subagent management on their own, and if instructed, can certainly set up a structured communication system for the group. Why these Herdr, Pi etc products on top?
Is Codex, Claude Code mismanaging when no tokens left? Or any other specific reason?
> Codex, Claude etc already do subagent management on their own
What about together? Can you reasonably do agent to agent interactions across harnesses? herdr makes that easy.
One of my favorite reasons for using herdr is what I'll call "everything is a pane". Running a new agent? Runs in a herdr pane. sshing into a system? Runs in a herdr pane. Anything that runs in a herdr pane, your agents can interact with. I will often do things like ssh into a system in one pane, and then ask pi in another pane to help me do diagnostics on that system. It never has to try to ssh in directly in the harness with my credentials, it just sends keyboard input to the pane, and reads the content of the pane. I can watch the interactive session as it goes by.
Another workflow that I have sometimes is I'll kick off some work in an agent, and instead of making a new sub agent, I'll actually make a new parent agent if I get worried that it's going down an unnecessary rabbit hole. I'll tell the new agent something like "periodically watch the agent in pane X, it is working on Y but I am concerned that it is losing track of its tasking. Periodically review what it's doing, and make sure it doesn't go off the rails. If it does seem to be going off the rails, give it guidance to bring it back on track" and that seems to have helped.
Subagent spends more tokens. It can be just separate sessions, just a new codex or claude session it doesn't have to be subagents that have to coordinate
It adds some abstractions and a plugin architecture for extensions. I checked it out but ended up not using it. I run my own harness of harnesses over tmux centered around tasks rather than sessions.
I suppose it is just the convenience of having an opinionated app that works with agents out of the box, tmux could be more powerful but similar to vim vs vscode, you need to tune it in.
It does read a bit like Claude was wondering around homeless in a foreign land and now has a native home. Good for him I suppose, immigration is not a fun process these days.
> Are there modern harnesses that don't do subagents/subthreads/whatever the same way?
At the end of the day it's a UX question. I could do the same as this (I use Paseo, similar thing, less obviously vibed copy in the docs if that helps...) with a bunch of terminal tabs and pi, and maybe some skills and scripts to help out, but it's a pain to manage.
> A new kind of agent workspace. Native and terminal first, with nothing between you and your agents. Every part of it is a plain folder your agent can rewrite, and every agent keeps running when the window is gone.
They're unironically echoing the infamously iconic 1980 Calvin Klein jeans commercial starring a perversely sexualized underaged 15-year-old girl:
"You want to know what comes between me and my Calvins? Nothing." -Brooke Shields
Except in this case, the romanticized "nothing" between you and your autonomous agents is a massive, multi-layered stack of legacy Unix abstractions, sockets, remote procedure calls, historic terminal emulators, escape codes, pseudo-tty device drivers, and modern infrastructure slop.
The reality of what is actually sitting between you and your agents:
- A herdr-core background daemon handling JSON-RPC IPC messages.
- A crates/ghostty-sys Rust FFI layer binding against libghostty-vt.
- A hardware-accelerated GPU rendering pipeline drawing terminal cell glyphs.
- A legacy Unix PTY subsystem providing master/slave pairs and terminal line discipline.
- An ancient mountain of VT100/ANSI escape codes and termios settings, with resize requests traveling through IPC and SIGWINCH signals notifying Unix terminal applications.
- A structured agent-status protocol carried by yet another terminal escape sequence, OSC 7501, then exposed through JSON-RPC.
"Nothing" says unfiltered, raw human intimacy quite like reading output from a pseudo-terminal file descriptor and watching \x1b[31m error logs cascade through an SSH pipeline and "pseudo" TTY device drivers.
Wrapping this tech in branding that sounds like a sultry, pseudo-erotic pitch is peak HBO Silicon Valley. TTY literally stands for teletype. We are supposedly building the frontier of autonomous artificial intelligence, yet our cutting-edge orchestration layer still routes interaction through abstractions inherited from noisy, electromechanical printing telegraphs of the 1960s. It's not a "native home for your displaced AI souls" -- it's an over-engineered wrapper shackled to the physical heritage of paper rolls and mechanical carriage-return timing: 110-baud Teletypes, separate carriage-return and line-feed operations, and terminal null padding conventions devised to give machinery time to move before the next character printed.
Didn’t know that about Brooke Shields, I’ll try to remove that phrase from my vernacular.
That said the rest of it is, well, complicated. Of course you can come up with something cool and new that doesn’t require all that plumbing, but how do you get people to use it?
If you build a beautiful piece of software that no-one uses, what was the point?
I switched to herdr this year after a decade of using tmux, and I haven't looked back since. Herdr is wonderful, not only because of the coding agent integrations, but also because it uses libghostty under the hood. So no terminal-emulator-inside-another-terminal-emulator shenanigans.
If you prefer GUIs, I can recommend herdr-gpui which connects to a herdr daemon. It has a bunch of goodies like remote phone pairing, and a built-in browser integration to send screenshots and notes to coding agents.
I've been using Zellij a long time and considered using Herdr, but the VC funding aspect scares me. I don't want to invest a bunch of time in a tool that's going to treat me as a product.
I used Zellij in the past, it is also fantastic and intuitive product. I had issues with... I think selection and copying and pasting which made me abandon it as I was getting frustrated. Herdr has been wonderful and I've been using it for a while.
Pretty much the first thing I tried to set up with I heard about Zellij and decided to give it a shot a while back was how to scroll back and copy paste text without using the mouse, since that's one of the core features of tmux for me. If my recollection is correct, the closest thing that I was able to figure out was some sort of built-in way to open output in a vim buffer, with the info I found when googling basically saying that it was intentionally not something the developers felt like it was important to implement anything closer to tmux for. Maybe it's changed since then but I distinctly got the impression that it using it would force me to adapt to a bunch of new workflows that didn't resemble anything I had seen before, and it didn't really seem worth the effort.
I don't know, this seems reasonable. I already know how to select, copy, and manipulate text in my $EDITOR. Why would a terminal multiplexer try to implement all of that rather than just let me open the terminal in my $EDITOR?
I'm not opposed to using the same shortcuts as editors (which is what tmux already does; you can even pick between emacs and vim shortcuts, in addition to configuring your own); there's just a lot of other stuff I _don't_ want from a text editor for my existing terminal output, like, editing text.
I recently switched to Zellij with Ghostty and the selection/copy/pasting seems to work just fine. I previously used zellij and kitty and I would have issues.
Same, but the cost of maintaining a fork fell through the floor in recent years. If a decent fork gains momentum, i can probably even stop spending tokens on mine.
But I’m already using ghostty. I installed herdr but made the mistake of launching it from within tmux and had no idea how to close it.
I then opened it somewhere else, but I had no idea how to interact with it. I also couldn’t get two other windows to “join” the same herdr session, so I gave up and went back to tmux.
For someone who has used screen / tmux for almost 30 years at this point, what does herdr do that tmux can’t?
I use Claude Code extensively. I also use plannotator extensively, which is how I found out about it.
It's all about agentic work. Tmux doesn't know what an Agent is. You use Pi, you use CC, you use Codex, ... they all have primary supports in Herdr with built-in integrations around each of these harnesses
But if you don't do any agentic work, Herdr has 0 value vs. tmux.
lol at the other replies waffling at you without describing the things that makes herdr appealing:
* agentic inbox [see which agents are waiting for your input/working/idle etc]
* agents can read, spawn, send-input etc other herdr panes
* makes cross harness orchestration / communication easier
fwiw I liked herdr initially but bounced after like a week because ghostty + notifications + harness extensions gave me enough that it wasnt worth learning the new thing.
If tmux can know what an agent is, it can be leveraged to communicate with an instance of that agent (input and output) given it owns the PTY, and that terminal output can be structured somehow.
If it can communicate with the agent, it can be used to drive the agent - to send it tasks, and communicate the status of those tasks.
And so Herdr - and Superlogical, and others, is a tool which leverages PTY ownership and terminal protocols to allow orchestration of coding agents across vendors using their CLI surfaces.
But Claude Code and Codex and the rest of tui harnesses are also very good at just typing in each other's panes basically when I need to send a message over. It's been working for almost a year now. I have been more intrigued by T3 code or something similar, just to be able for harnesses to message across machines.
overall not so much, but basically if inside tmux you have 100 terminals opened and only 10 of those has 'agents' it will allow you switch quickly between those 10 terminals without spending too much time on locating those. also it will show statuses about agents, is it work done already and you may need to check it and stuff like this.
I would assume other models have their own harnesses as well but I guess if you want it at all in one place that would make sense. I just need a single model.
The point about libghostty is that Herdr doesn't try to reimplement shell forwarding or emulation, instead it uses the same code your terminal already does. This avoids compatibility problems with more complex TUIs.
Running `herdr` always opens the same session, the app is a just a UI for a server running in the background. But there is no reason to open it in multiple windows, the point is to manage all shells from inside it.
Then you can from another machine do `herdr --remote xxx.local` (over Tailscale) to access the same session, or use `herdr machine add` to get your shells from all machines at once - I don't really use this part.
From someone who also used screen and tmux for 30 years: 3 reasons why I switched to herder. They may not be relevant to you and in that case tmux is just great. One: visual indication which agent is working and which is waiting for an input. Two: native support for remote herdr sessions - in the same view. Three: it’s responsive and can be completely touch-driven through menus on a small iOS screen in Termius.
I was looking into Herdr too, but could not bring myself to switch away from tmux yet (there are a couple of shortcuts that I have in my workflow, which I can't config in Herdr).
What I ended up doing was to use the tmux agent sidebar, which, with mouse support, I think does a lot of the same things already.
All of this, plus it knows what agent is running in which windows and what session that agent is working on. If you have to reboot, it can reconstruct all your terminal setup and restart the agents on the right session automatically.
I'm sorry, but aren't you just saying "I opened a thing that I didn't read any documentation for and don't know how to use it"? Did you know how to use tmux without reading the docs?
> Herdr is wonderful, not only because of the coding agent integrations, but also because it uses libghostty under the hood. So no terminal-emulator-inside-another-terminal-emulator shenanigans.
Now I’m interested. Does it have basic feature parity? How does it compare to tmux in your experience?
I was able to have a language model configure herdr with the same bindings that I was using in tmux.
It was a very easy transition from one to the other.
I started using it last week, the tui interface is very polished and it is mouse-friendly. It also allows you to configure the notification channels. Its a different beast from tmox, I'd say its quite more user-friendly and gives you a list of your running agents and their status directly on the interface.
herdr is by far the most polished tool in its category. Software geared towards agents is usually vibe coded -- and you can tell. There will be bugs and edge cases and it feels like no thought/care was put into it.
herdr doesn't feel that way to me. The marketing/landing page definitely has LLM-isms, but the application itself is very stable and well thought out.
A true UI is the next logical step and they seem to have done the obvious things. The one surprising bit that nobody seems to be talking about (yet) is the entire UI looks like its controlled via "recipes", so you can customize the UI in ways that you couldn't with traditional software: https://herdr.dev/layouts/
I hope this means an official mobile app is on the horizon!
> herdr is by far the most polished tool in its category.
Can't speak to this, but I like that I didn't feel like I needed to try an alternative. It just did exactly what I needed it to do.
That said - I couldn't find a way to move a pane to a new tab. I can swap panes within a single pane (select -> right click -> swap with focused pane), but I can't find a way to do it between tabs. Maybe it's a skill issue. Anyone have any ideas?
Although this isn't a direct answer to your question, but everything that can be done in Herdr is trivially discoverable by asking a CC (or equivalent) session inside Herdr. In fact a lot of the things I do in Herdr I do by asking an agent to do it for me, instead of trying to learn how to do it manually.
Some things might require a non-trivial workaround, but a coding agent will also happily build and test it for you.
Playing with this for a bit, the customization you can do is really cool. You can spin up an agent, tell it to add some functionality, and it all reloads instantly.
For example, I wanted to be able to hide chats from providers when my usage is out for the 5h/weekly period
I'm not sure how they're doing this considering it seems to be written in Rust
Agreed. I was really sceptical of it and was in the camp of "tmux does all I need" but it is honestly fantastic and ever since they added the ability to access multiple ssh machines from one herdr session I just live inside it.
Have you tried T3 Code? IMO it goes a step further on UI by dropping the idea of interacting via a terminal. Also it has had an official mobile app for months.
That's funny, I got tired of tmux the other day (it got into 100% spin loop from an unsigned int underflow used as a loop condition variable. and I debugged it all by myself like a big boy, without LLMs!) and built something that seems to be identical in architecture in terms of the "supervisor which allocates pty, just stream bytes, client decodes" part. I even use libghostty for the HW accelerated GUI I built. It's the right way to go imo and it's not too difficult to build... hence...
I don't really understand why they would turn this into a company. I become suspicious of products that raise a lot of VC money, implying that they expect a unicorn-ish valuation, where I don't see them being able to justify their valuation via obvious monetization techniques (e.g. selling a classic license for a GUI app where you get updates for some period of time, usually a year, and then you get to keep the software, but no longer get updates, just in case people forgot how things worked before subscriptions).
That usually implies something like selling cloud subscriptions, vendor lock-in, or selling data. I'm happy to be wrong, but it's just an extremely common pattern. And as an aside, I know there's an impulse to monetize everything, but I think we should probably fight that impulse.
Different philosophies. herdr has defined a few core primitives (machine -> space -> tab -> pane -> terminal) and lets you customize all the ui around it. Among all the ADEs I've used this is most reminiscent of emacs.
Orca is a bit heavy in asking you to define projects / worktrees up front. Anything you do, even quick edits starts with a worktree. Projects get their own modal and file browser before you start working. Herdr just lets you cd, mkdir, get going. But it's also worktree aware and organizes them under your repo when you do create them.
Herdr looks awesome, exactly what I need. However, automatic updates worry me in the light of recent supply chain attacks. I don't want be in a situation where one day I just find out I'm running malware on my computer. I'm not sure if there's an option to turn it off.
Darn, title had me thinking they released a mobile app to compliment the TUI daemon.
termux+mosh is fine, but especially on a phone I'd love a slightly more intuitive (or really, efficient) UI. Especially as mosh breaks touch/mouse input on the remote host.
I've never understood the point of the desktop GUI harnesses to run TUI (or at the very least, primarily text-based) programs anyway, but perhaps they're better for onboarding less tech-savvy users.
Moshi is pretty fantastic. I’ve been using it for several months and it works great. The developer is also launching fixes and improvements every week. It’s gotten materially better over the time I’ve been using it and it started out pretty great.
Looks good but not way I'm paying a subscription for a tool like this, make me pay a one time fee and give 1 year of updates (90€ would still be too much for me, 50€ I might reconsider).
Ah sorry didn't open my eyes apparently, still a bit expensive to me but I will probably test drive it then, hope the desktop app is good it's probably even more important than the mobile app to me and does not seem like their focus at all.
I'm using Ovis a plugin for herdr that adds a web app wich works well enough on web and desktop, but I feel like I'm playing Russian roulette every time I'm updating it and there's updates every day... It's the trust issue of vibe coded apps... Moshi I would feel slightly more confident they won't voluntarily add malware to it and pay attention to security.
I got access around a week ago but found it quite unintuitive and found myself sticking to herdr for now, but was probably a bit quick to judge. What makes tern great for you?
Having used herdr since its release, the recent 'Emacs is 50' made me decide to go back to it and just vibecode herdr features i use into an emacs package.
It works great so far. I can just ask it to update its layout whenever i think something better is possible. I wont be going back, but the one thing that might change that is a mobile app.
this new herdr app is actually very close to emacs in philosophy. Instead of the C core and lisp runtime herdr has GPUI (rust gpu accelerated UI drawing) and herdr-core which is the agent daemon. There's some vocabulary for spaces tabs panes agents, and js sits on top to define UI/UX. Very little is hard coded, and default app functionality is implemented in the same way that users can customize, and changes hotreload without restart. Almost everything seems to be just js/json just like most emacs is just elisp.
herdr is moderately well known if you look into agent multiplexers. IMO it is the best one from what I have tried (Paseo and Conductor are competitors in this area)
It's not a harness -- it's really more like a tmux for agents.
Your comment is so comically dumb.. Herdr is yet another recently vibe coded ai helper in a niche category. You wouldn't miss anything about frontier ai or coding agents if you don't use herdr and/or terminal multiplexers in your workflow.
I have been using herdr for a few months. Then i bought a new machine, installed claude code and asked it to setup herdr. Shockingly claude said it has no idea what herdr was. This was weird.
The Herdr website [1] has one of the best demos I have seen, one that I think would blow away those who aren’t nearly as AI-brained as I am. Why are we locked into old, boring UIs when I can just ask the agent for something like cows in a pasture or Windows 98?
I mean one where you can say to herdr, make that look like Windows 98. Still can’t think of a good example that’s not already from Windows 98 unfortunately.
News to me, I guess. I have never seen anything like these demos. Most themes in software I use change colors or rearrange existing UI elements rather than create kanban boards and floating interactive cows where they did not exist before.
Don’t get me wrong, I think having much more customizability now that maintaining complex configurations is easier is absolutely the direction we should be headed in. I don’t think floating cows are going to be the ultimate goal, but it is a step in that direction.
But this sort of theming specifically is something that was experimented with a couple decades ago and given up because people wanted simplicity and consistency instead. It’ll be exciting if LLMs reverse the trend.
I loved Herdr at first and got a massive productivity boost. Installed a bunch of plugins and created a few myself, but felt a bit messy after a while. When working on 10+ tasks I had a hard time getting a proper overview of the agents and high cognitive load when swapping between the tasks.
I built my own app based on what I liked with herdr and added some other things, but with this new herdr update I might switch back
https://svall.dev/
Still a work-in-progress[1], but I have a mobile app for herdr also built with a native GPU interface (but not using gpui), optional LLM usage (outside of the agents themselves), designed for a hands off voice-enabled approach to development. The STT/TTS are both on-device. Yes, I used agents to build it, and you can see the development record in tree under docs/design.
I love the "What we send" information - I looked that and figured "that's not too bad - fine, have it!". I think this is the first time I've seen this kind of transparency.
Telemetry does not include any data from your terminals, or names of files, folders or projects. In fact they even provide an example JSON payload for you to inspect in the settings dialog.
Supports SSH to remote machines out of the box, cool. I wonder if the protocol is more efficient than the TUI over SSH, would be nice on crappy connections when traveling.
It is, but if you send more efficient deltas, batch packets more intelligently, and introduce non-blocking UI tricks, the experience can be improved by quite a bit. mosh demonstrates some of that.
If you like Herdr, you might also appreciate https://github.com/BrokkAi/mjolnir/, which is similar in its multiplexing features (although built on ACP) but offers cross-provider subagents, memory and session log sync, live migration of sessions across profiles and providers, and load balancing across multiple subscriptions.
Does anyone have experience in how herdr compares to Zed?
I'm currently using Zed as my main "agent orchestrator" and IDE because it handles various simultaneous agents quite well, using different harnesses. I'm especially a fan of the diffs for code review and the ability to easily highlight code and add it to context. The way it handles worktrees and git is also nice, IMO.
I have had much better experiences with the desktop apps for managing LLMs compared to the terminal.
Tasks like switching between multiple sessions, copying and pasting, viewing diagrams/web pages, and resuming remote sessions I have found significantly more polished when using the desktop app equivalents.
I don't think I need it, but Herdr has been absolutely rock solid for me. And I have this app Pairfob that I can check things up on the phone if I really want to. And while I don't use it quite as often, seems to be more solid then Chatgpt (Codex) and other apps.
I have 2 machines and laptop linked together and running things have never been easier.
I left tmux+vscode a few months ago and haven’t looked back. Clearly this new thing is better than flipping between a dozen tmux panes and other windows.
But I’m using my own “IDE” inspired by herdr and pane, worktrees down the left and panes on the right, but besides terminals the types of panes also include:
- A native Obsidian-level markdown editor with support for comments that agents can see
- Claude code with chrome for context size, and a custom UI that looks more like the web
- A diff viewer with support for comments
- A browser
- Can tag worktrees with PRs and issues
- Everything (except the browser) works seamlessly over ssh
It’s all zig+gtk, all vibe coded. I rarely open vscode these days.
Maybe the herdr plugin ecosystem can support all this too. I kinda want to get out of the business of hand-polishing UI.
I actually did the same but at some point was like "this is good enough" and it works well. Diffs, comments on diffs, integration with Linear and worktrees is good enough for now and gives me a productivity boost. Couple of other apps in the space but at the time none supported Claude Code in terminals
I recently searched for an application to manage my agents and work streams, on top of Claude Code. Chat called them "task cockpits", but I'm sure I've seen other terms used for this category. I found a bunch that were abandoned already. And I didn't see Herdr until I already selected Pane, and committed to giving that one an honest try. It makes me wonder, is there any winners in this space yet, or is it just as fragmented as it seems?
I can’t seem to understand if this is something that could also be used without agents. I see the appeal of having first support for agents but I’d also like something I can use myself like I use tmux, without agents.
If agents didn’t exist, would herdr provide a nice UX as a terminal multiplexer?
If you like the idea of herdr, but want something that isn't opinionated on terminal vs. desktop app and has mobile support, you might like https://github.com/codealong/codealong
Nice! I have been using herdr for a few months, love it and for me it is a step up from years using tmux. The new native Herdr app is yet another step up. I use multiple Macs with Tailscale connections and the Herdr app makes jumping around to different projects and different machines easier.
I was wondering this, too. The desktop appears to be based on herdr-core, and naturally I'd have imagined Herdr Server (terminal) to migrate to herdr-core as well. In fact I'm not even sure why they built herdr-core when they had Herdr Server ..
I was initially excited because I thought I could just try out a new frontend while keeping my 10 existing spaces each with 10 worktrees each with 2 or more tabs untouched.
Haven't used Herdr before - was looking at README and was looking to check screenshots and documentation. However, Chrome reports the SSL cert to the site https://herdr.dev as invalid. Will wait till it gets resolved.
It works now. Haven't restarted browser/mac - so not sure what went wrong.
Anyways, Herdr looks powerful - will try it soon. One thing I find missing in every single one today, is that they do not allow using a different AI account based on the workspace/project you are on. It would come in handy in case of shared family machine.
I don’t think they need to “allow” this specifically. Every new pane opens in a shell prompt. You must start the agent from there, so you can start each agent with different keys/accounts already.
> The goal is to link your machines with a single command and stop worrying about how to reach them. Herdr Cloud would provide the middleman layer for those connections.
Right now the new herdr has more ... just more. Rex has the polish, attention to detail, and attention to performance.
They both follow the same "gui with a cli to manage the engine which owns the PTYs, has some interaction protocol, wraps coding agent interfaces, and you can connect to remote machines" model.
Unfortunately they are both GUI only atm and so I'm still using herdr-tui over ssh until it breaks or we get a tui surface.
I love the current model, where I can use the tui in ghosttty or wezterm or iterm, or pop an ssh session, and if it makes sense there's a gui version for whatever device I'm using. It's like magic, I'm free to do whatever from wherever. Forcing a GUI feels like them asserting ownership or control. Now I need their mobile app if i want to develop remotely, etc. I recognize the irony, the core is open- I'm free to develop against it - but we all know that realistically only a handful will and those will lag. Also i am sure that GUI is first because that's where the users are. Eh.
Didn't have much time to try it this morning, but so far the Herdr app looks fantastic. I am already using herdr on the console to control Claude on my Mac Mini, but having to log in again after closing my Macbook was always annoying. And I missed seeing the code diffs and being able to commit changes in VSCode, which is still my preferred development environment today (with the Claude VSCode extension).
But Herdr app addresses both issues. I don't need to reconnect, and it has a diff tool that allows me to review and commit. Looking forward to work with it!
This new GUI looks awesome, however it's not clear how to do the agent integration? With the TUI I installed the skill via `npx skills add https://github.com/herdrdev/herdr --skill herdr` and the agent knew what I was talking about when I said herdr, now I have to tell the agent 'Read `herdr-app --skill` and ...' - seems like extra steps?
Is anyone else finding that it doesn't pick up Claude Code agents? My Agents pane stays empty no matter how many instances I start and work in. Appears to continue to work fine in herdr, though.
Not the new one though. I'm sure it'll come. The new herdr is not backwards compatible, and the TUI is unlikely to see development (at least in its current form).
herdr is great, and the original runs-in-a-terminal vision was cool, but at least on OSX it has a ton of selection management bugs that are pretty sucky.
As I understood, this is a completely different thing from what we had so far (herdr TUI), fully incompatible - so technically the only common thing between them is the team and the name.
It doesn't see the sessions in my already-running remote herdr server, you cannot access the Herdr App remote sessions from your mobile phone (previously you could via SSH client like Termius).
I was actually kinda pissed when i realized this. Then i got an invite to Rex from superlogical, and I was free! Only to find out it's the same damn thing :)
Heeler has a PR open to target the new core, but no sign anyone is working on a tui for it :(
Which of these various new multiplexers (herdr, cmux, etc) work over ssh, and which have protocols underneath?
I've been scripting tmux for a while now, and it's not the greatest. And when I have a bunch of opencode sessions rolling along often my tmux could usage is smoking. I'd be happy to try other things. But I'd way rather get into something with protocols, that is open to play.
Hydra might be worth a look. The daemon owns the various agent processes (any that speaks ACP), and everything else is a client of open interfaces: a REST API for managing sessions, and ACP over a WebSocket for attaching to a live one. The TUI, a browser UI and VS Code all use those. So you can script it with plain HTTP instead of tmux. It doesn't use ssh; remote access is TLS. The TUI shows one session at a time, and ^P jumps between them.
Not having a normal installer/uninstaller is a weird trend, I dislike these powershell scripts that dump stuff all over and have no way to uninstall it after.
Especially when it's an AI agent tool and an AI agent can probably make an EXE installer in a few minutes.
I just had a look at the site and screenshots, I usually run my coding agents containerized (one per project), anyone else doing this and wiring it to herdr? is it possible?
from a quick glance it seems to me it installs cc/codex/opencode/etc on it's own and just does multiplexing
I have a project that auto-containerizes with its own frontend, pretty sure you could wire the backend to herdr but haven’t done it yet: https://yaac.dev/
at the end of a day I started to one-shot some simpler workflows; e.g. I already have various scripts for spinning work-trees, setting up deps, picking up a clearly defined issue, waiting for some delayed CI checks, doing reviews with different agents, workflow waiting for another workflow etc. - without passing context between
depends on what you're looking for - for whole worktrees setup I use worktrunk [1] which handles all stuff like spinning up deps, containers, envs so ports do not clash, every worktree have clean slate so local tests work etc. solves like _a lot_ of problems I ran with worktrees in real-life apps
I started using dagu fairly recently so the rest is kinda a ball of various ruby scripts at this point so I can't give any good pointers here as I have no idea yet xd but overall I have some deterministic stuff (pick up some issue, setup worktree first, resume if crashed, implement, review and so on) build around just running claude/codex (one-shot)
there are _tons_ of solution for orchestration so it's really hard to choose from tbh; I gave herdr try this morning and ported my scripts so it works with their desktop app - looks kinda nice I guess; dagu have scheduler as well which is cool, but connecting different machines to single herdr server also looks kinda fun... so I guess I will continue experimenting until then ;)
Looks good but doesn't have a browser yes, like all the desktop applications of major harnesses have. Having a browser accessible to both the user and the model is extremely useful. I know you can do it via MCP but still, having a baked in one is better.
It's hard to beat Claude Code with /remote. I can easily keep things moving from my laptop, phone, or another desktop. I realize Herdr can technically do that too and across more agents, but it's more to setup and I really don't need that.
I also built a similar thing a many. And in my experiments (even though iroh is great) iroh took significantly longer to reconnect vs mosh and ethernal. And actually plain old autossh was the best XP for me to close laptop get on traincopen again on mobile Internet.
So eventually I removed everything other than autossh.
Seeing as how many people are building similar things I'd love to share more. Maybe wanna ad each other respectively to the repo's ?I can share some of my benchmark results
For me connect time is not really a concern, it happens once when I open the app. In my project I'm wrapping agents using ACP and presenting an actual GUI, so latency doesn't really have a perceptible impact. Running Claude Code in a console over any kind of remote connection is gonna feel bad.
Rather than for network performance I'm using iroh because I don't want to need a direct network path to the machines running my agents. Yeah I have Tailscale but that's more work to set up and I don't want my app to depend on users have a TS network.
To elaborate on my usecase that herdr didn't so at all the the beginning but still does not do well: I have many vm's in different clouds working on different projects with different secrets that I want to keep isolated ( now also at home lan machines which when im travel get aliased to their tail scale address)
So my biggest requirements are around making isolation between environments convenient but safe from supply chain attacks .
+1 I tried quite hard to make things work with tmux, zellij, and even some light custom stuff, but if you have multiple agents running on multiple boxes herdr just works unreasonably well. I really didn't want to like what seemed like a random mix of features, but they've stabilized it to a sensible capability set.
Glad to see the desktop app and after did some research seems its like rex and with some once feature and it even have a Win98 easter egg if you can find it:)
Currently I am not switch to it because I just have my own app talk to the herdr daemon via my apps builtin tailscale and it works great
The app doesn't even work as advertised. Opened the app, started a new session, closed the app, and my session was gone. I thought this was supposed to be persistent sessions. I think i'll keep using Cmux.
Hmmm I tried it, but it didn't detect my open herdr sessions in the added machine. I don't see much benefit over herdr in Terminal/Ghostty, and I need to reconfigure my agents to make it work too.
Anyone using this or similar with self-hosted models/engines (not on local machine)? Didn't immediately see anything about that anywhere in the documentation.
Fantastic news, I may have to switch back (or combine this with Paseo on my phone). I hope the RAM usage is a lot better as Paseo tends to OOM now and again (the desktop app).
The tabs mechanism didn't do it for me so I rolled my out (self-plug at [0]).
While it's not "agent-first" thanks to rich MCP I often use direct terminal control from one orchestrator agent to drive another one, and I haven't seen any generic solution like that.
Since I'm moving my work VM yesterday I even started working on ssh persistence using planted cross-compiled PTY host library. If anyone's interested check it out, I expect to roll it out within a week or so.
In its current incarnation, superlogical is exactly the same as the new herdr, with the same gaps. A TUI does not appear in their backlog either. They are a native app.
Yes, but Rex is built on libghostty with the multiplexing and agent integration built in, built by people I trust who don’t let the LLM write all their docs.
My cat decided to sit on the power button of my laptop, and upon restart herdr-app greeted me with empty terminal panes, whereas herdr-tui was just fine.
And seems like they use "herdr-core" (hc) to actually run sessions instead of herdr server, that flushes events only every 100ms -- the agent output is quite jittery with that.
I use herdr daily. One thing I’ve always missed from all of these solutions, though, is the ability to actually make code changes..none of them seem to offer that properly. Am I the only one? Is the goal to abandon code entirely?
feels like they're trying to get ahead of superlogical's rex.
took all of installing it on windows and setting up my mac as an SSH machine to see the slop all through the UI. i'll just wait for rex's windows client i guess.
I really enjoyed herdr at first but ever since they got VC funding I am avoiding it. Enshittification and cloud features due to vc/monetization pressure incoming.
I guess there is some nuance here. Would I pay for a very good agent multiplexer that makes me a lot more productive? Sure, same reason I pay for other apps or services.
But for Herdr specifically? Probably not. There are tons of similar free alternatives and even vibe coding your own is very easy with all the open source code of both herdr and all the alternatives available.
> Would I pay for a very good agent multiplexer that makes me a lot more productive?
That is not the question I am asking.
> But for Herdr specifically? Probably not.
So no then.
That is why they went for VC money since the author knew that donations weren't enough to maintain it in the open.
> There are tons of similar free alternatives and even vibe coding your own is very easy with all the open source code of both herdr and all the alternatives available.
Why would I waste tokens vibe coding something that would go unmaintained in 6 months (because it is "free") over a properly funded project that will last much longer than that?
You dont have to vibecode and maintain it when there are 20 other projects that do the same or similar thing. And maintenance is not free, but you know waht's even more annoying? Getting locked into a product that is pressured to achieve VC returns and thus becomes enshittified. There is no moat here and it's painfully obvious that open source will win in this category in the long term.
> Getting locked into a product that is pressured to achieve VC returns and thus becomes enshittified. There is no moat here and it's painfully obvious that open source will win in this category in the long term.
Herdr is...open source? and is also compatible with the free software principles. So where is the "lock in"?
You have not answered how will Herdr (and others) fund themselves to maintain the project without some form of sustainable funding without VC funding?
Not that I support VC funding per se, but without meaningful donations to support the maintainers, this is usually what they end up doing.
Nope. I think you are confused since Herdr was free software and licensed under AGPL 3.0 [0] before it was VC funded.
One (emacs) gets millions in donations via the FSF and the other get little to no money or donations for being "free software" from the start despite both of them being "free software".
2004 called. They want to return their disemvoweled penultimate "e".
2004 - The Big Bang (Flickr) - Co-founder Caterina Fake wants to buy Flicker.com, but an offline beer brewer refuses to sell it. Out of options, she suggests dropping the "e". It works, and Flickr is born, instantly setting a precedent that typos equal modernism.
2005–2006 - The YC Copycat Flood - Paul Graham launches Y Combinator, and the application form might as well have had a checkbox for "did you delete a vowel to look disruptive?" Scribd (YC S06) drops the E because buying a real dictionary noun requires actual revenue. Twitpic and Pixlr emerge to capture that sweet, sweet pre-seed funding. Even Twitter originally launches in 2006 as "Twttr" because the actual domain is owned by a bird enthusiast. (They buy the "e" back six months later once they realize they look ridiculous).
2007 - The Peak of Style - Tumblr launches, proving the vowel-less trend isn't a passing fad but the official language of the new internet. Dropping the "e" becomes shorthand for "we have venture capital, a minimalist mobile interface, and absolutely no monetization strategy."
2009 - The Aggressive Pivot - Grindr launches. The convention branches out of pure utility and enters cultural lifestyle apps, cementing "-r" as the universal suffix for a location-based mobile service. Every YC batch now features at least three "Uber for X" startups named after a mangled verb.
2011–2012 - The Saturation Point - Companies like Blendr and Fiverr continue the trend, but the public begins to push back. Every SaaS tool on Hacker News looks like a typo. Critics and users start pointing out how incredibly annoying it is to vocally explain a business name ("It's called Taskr... no, without the E").
2015–2018 - The Death Kneel & Corrections - The joke gets old. Brand consultants begin labeling creative misspelling a "cardinal sin". Startups realize they are losing millions of organic visitors to the properly-spelled domains owned by domain squatters. The internet enters the "-ify" (Spotify, Shopify) and "-ly" (Bitly) eras, completely abandoning the broken suffix because pg wrote an essay saying missing vowels hurt your SEO.
2019–Present - The "Adult in the Room" Era - The trend is officially dead. Modern YC startups pivot toward ultra-literal names (The Browser Company) or aggressive, serious abstract words (OpenAI, CoreWeave) to pretend they are enterprise-ready AI infrastructure rather than a wrapper wrapper wrapper.
Postscript - The Ultimate Regression - The convention reaches its absolute, braindead rock bottom with Elon Musk. After burning 44 billion dollars to buy Twitter just so he could unban literal Nazis and amplify white supremacist conspiracy theories to the front page, he rebranded the entire platform to "X.com". It turns out "X" is just "Ex" without the penultimate "e" -- a fittingly hollow, single-letter monument to a sociopathic tech bro who couldn't innovate a functional name if his taxpayer subsidies depended on it.
Herdr is truly awesome. Game changing almost. If you think of all the overly complex agent setups[1] and overly complex software factories[2] to write code... herdr and a pick-your-coding-agent replaces all that with a very simple one thing which just works.
There are many ways to do what herdr does, like tmux, and rex, but in a nutshell:
You launch cli things (claude code, cli tools etc) anywhere inside a herdr process. On your own laptop you magically stay connected to all of them and can monitor, send commands etc. And those processes can talk to each other and control each other too if u want!
For a single person with herdr+tailscale, you can make your own evolving software thingamajig across all the possible computers u have access to (pls don't call it a factory, we didn't call your fingers a factory in the dinosaur days did we).
p.s if someone from cloudflare reads this, would love to have a herdr process in a worker so it appears on my laptop as a thing i can do stuff with. while idle it goes dormant, i send a keystroke and it wakes up, stays awake for a bit.
p.p.s herdr team if u read this can you have a built in so i can instantly launch a vm in one of the many provide a instant vm (or sandbox or whatever the cool kids call it these days). I have been using the railway TS sdk to quickly spin up and throw away sandboxes, but why not just integrate <choose-your-sandbox-provider> in herdr and that is one way to charge money - pay herdr or provide my API key for vm/sandbox provider. Bonus points if it can sleep and charge less. Sort of like fly's sprites but more awesome.
Because my job has multiple servers with different needs and tooling, and I must do my development within the appropriate environment.
I'd also _much_ rather do my development on a persistent, always-on host so I can switch what computer I'm using for a head or turn off my computer and drive home while Claude works overnight.
I have multiple projects running in different locations/timezones - so I don't often need to have an agent running on a pc elsewhere - but just being able to smartly use herdr (also using rex) to set up a persistent service on those computers removes all the friction which used to exist with ssh keys and other things.
So I could do all these things before, tools like herdr make it easier, smoother.
Now tmux and ssh keys enabled all this before. So in a sense, nothing new here. but its a bit like dropbox - i used rsync back then, dropbox just made it easier to setup sharing and use without thinking.
My current anecdotal example: if you're building cross-platform native apps, you need to run agents on macOS and Windows, because xcodebuild only runs on macOS and the Windows SDK is only available on Windows.
> A native home for Claude Code, Codex and every coding agent. Every pixel is a file your agent can change. Close it, they keep working.
First two sentences give zero information what it does. Third one is useful but i think many harnesses do it?
Continuing:
> A new kind of agent workspace. Native and terminal first, with nothing between you and your agents. Every part of it is a plain folder your agent can rewrite, and every agent keeps running when the window is gone.
Okay, more of the same. Let's press on.
> Your agents run agents. And hear back.
> An agent hands work to helpers and keeps going. When they finish, their reports land in its input. No copy-paste between windows: herdr-core carries the messages.
Are there modern harnesses that don't do subagents/subthreads/whatever the same way?
---
There are some interesting bits but the vibed copy is for me offputting and a sign even the author might not understand what he made.
I've been using this for a few days now and it's absolutely fantastic.
Yeah that copy is pretty bad. My understanding (someone correct me if I'm wrong), but the herdr native app is an improvement over the existing TUI multiplexer, with a new API for facilitating comms between agents. Herdr is really useful if you've not tried it. I'll be giving this a go (I hadn't really been using agent comms between panes anyway, more just for workspace navigation). Seems you can think of herdr as a 'harness for your harnesses'!
What's useful about Herdr compared to tmux and claude code or codex's built in sub-agents?
I can't speak to the GUI, just the TUI compared to tmux.
herdr is very flexible, but it shines when you are working with multiple projects, with multiple worktrees, and multiple agents within those projects.
1. herdr uses workspaces to help organize tabs. Conceptually, a workspace probably maps to a project, but it doesn't have to.
2. herdr has native worktree support, and can create a nested workspace per worktree with a keyboard or CLI command. It can check out worktrees for you
3. herdr identifies agents across tabs and workspaces and pulls them out into an extra navigational pane. It shows the status of those agents (e.g. idle, working, waiting, done), and can play a sound when they change state. This helps you track multiple agents across multiple projects and quickly switch to them when they need help without needing to remember "claude is running in space 2, tab 3"
4. herdr integrates with your agents, which can control herdr's UI and rename tabs based on their status.
5. All this is scriptable via the herdr CLI, and agents are very good at it. But it also has full mouse support, if you're new to multiplexers its very easy to start interacting with it.
I usually keep one tmux session per Python worktree, with a local Postgres test container in another pane. The status view sounds useful here; does Herdr detect waiting via agent hooks or infer it from terminal output?
Uses hooks for some, output for others - depends on the harness in the pane: https://herdr.dev/tui/docs/agents/#supported-by-herdr
May depend on harness, but for claude/codex its currently inferred from terminal output.
> What's useful about Herdr compared to tmux and claude code or codex's built in sub-agents?
Agent status detection. Single view over local and remote agents. /herdr skill that lets agents list other agents, peek into their panes, send keys (yes, you can make a lot of it work with tmux and some elbow grease, but herdr provides nice cli/api for it and works out of the box) including to remote ones.
In my mental model, subagents are also different from top level agents in tabs: subagents are only great as temporary workers you never want to interact with.
It also lets you message between Claude Code and Codex sessions. If you're using Claude Code, their built-in messaging system only can see/message other Claude Code sessions (for obv reasons).
This is an unlock, because these companies will (prob) never build this feature.
[dead]
I think i built something similar after herdr was released but i can say the reason is at least
tmux is definitely usable but moving between workspaces/repos can be annoying if youre managing mahy many things at once. and difficult to organize. more over having agents manage other agents via tmux can lead to garbled input/output
i havent used codex. claude subagents i think arent the right comparison - the closer comparison would be claude managed agents. but i imagine one reason herdr might be better would be that you’re not stuck in the anthropic ecosystem
If you prefer tmux orchestration you could give my plugin a try.
It's just a few bash scripts enabling agents to message each other. Works with the desktop apps as well so you could have a Codex or Claude desktop app orchestrate tmux sessions.
https://github.com/notaharness/plugins/blob/main/orchestra/R...
Integrated experience. You can connect to multiple machines and access terminals on them through spaces in the left nav. Agents across all spaces are listed in a dedicated section for quick access.
Each space is persistent contains persistent terminals like tmux, so disconnections do not affect them.
The original tui-app is more stable. I’ve had instances of my agents randomly getting killed on herdr-core + macOS app stack.
I'm also curious, if you just ask Codex, Claude etc already do subagent management on their own, and if instructed, can certainly set up a structured communication system for the group. Why these Herdr, Pi etc products on top?
Is Codex, Claude Code mismanaging when no tokens left? Or any other specific reason?
> Codex, Claude etc already do subagent management on their own
What about together? Can you reasonably do agent to agent interactions across harnesses? herdr makes that easy.
One of my favorite reasons for using herdr is what I'll call "everything is a pane". Running a new agent? Runs in a herdr pane. sshing into a system? Runs in a herdr pane. Anything that runs in a herdr pane, your agents can interact with. I will often do things like ssh into a system in one pane, and then ask pi in another pane to help me do diagnostics on that system. It never has to try to ssh in directly in the harness with my credentials, it just sends keyboard input to the pane, and reads the content of the pane. I can watch the interactive session as it goes by.
Another workflow that I have sometimes is I'll kick off some work in an agent, and instead of making a new sub agent, I'll actually make a new parent agent if I get worried that it's going down an unnecessary rabbit hole. I'll tell the new agent something like "periodically watch the agent in pane X, it is working on Y but I am concerned that it is losing track of its tasking. Periodically review what it's doing, and make sure it doesn't go off the rails. If it does seem to be going off the rails, give it guidance to bring it back on track" and that seems to have helped.
Herdr doesn't force any particular workflow, it's just a way to organize terminal sessions with built-in agent awareness
Subagent spends more tokens. It can be just separate sessions, just a new codex or claude session it doesn't have to be subagents that have to coordinate
For me the biggest thing is subagents are harness agnostic
I used to hit usage walls with CC/Codex harness using native subagents
With herdr I delegate some work to another herdr pane and it runs Claude code, codex, opencode, pi, whatever model is appropriate and has usage
It adds some abstractions and a plugin architecture for extensions. I checked it out but ended up not using it. I run my own harness of harnesses over tmux centered around tasks rather than sessions.
I suppose it is just the convenience of having an opinionated app that works with agents out of the box, tmux could be more powerful but similar to vim vs vscode, you need to tune it in.
Compared to the usual agent manager guis, it's not even that opinionated.
The automated session resume/workspace restore after restart etc is quite nice, and copy/paste is much better than tmux.
My question also!
Had a play with the native app - it's a complete new set of keyboard shortcuts from the TUI. Don't really see the advantage.
Quite popular theses days, I imagine is just another competitor in the same field as T3[1]
[1] https://t3.codes/
It does read a bit like Claude was wondering around homeless in a foreign land and now has a native home. Good for him I suppose, immigration is not a fun process these days.
It does sound like that now that I think about it.
> Are there modern harnesses that don't do subagents/subthreads/whatever the same way?
At the end of the day it's a UX question. I could do the same as this (I use Paseo, similar thing, less obviously vibed copy in the docs if that helps...) with a bunch of terminal tabs and pi, and maybe some skills and scripts to help out, but it's a pain to manage.
> A new kind of agent workspace. Native and terminal first, with nothing between you and your agents. Every part of it is a plain folder your agent can rewrite, and every agent keeps running when the window is gone.
They're unironically echoing the infamously iconic 1980 Calvin Klein jeans commercial starring a perversely sexualized underaged 15-year-old girl:
"You want to know what comes between me and my Calvins? Nothing." -Brooke Shields
https://www.youtube.com/watch?v=GiQUMypDNoE
Except in this case, the romanticized "nothing" between you and your autonomous agents is a massive, multi-layered stack of legacy Unix abstractions, sockets, remote procedure calls, historic terminal emulators, escape codes, pseudo-tty device drivers, and modern infrastructure slop.
The reality of what is actually sitting between you and your agents:
- A herdr-core background daemon handling JSON-RPC IPC messages.
- A crates/ghostty-sys Rust FFI layer binding against libghostty-vt.
- A hardware-accelerated GPU rendering pipeline drawing terminal cell glyphs.
- A legacy Unix PTY subsystem providing master/slave pairs and terminal line discipline.
- An ancient mountain of VT100/ANSI escape codes and termios settings, with resize requests traveling through IPC and SIGWINCH signals notifying Unix terminal applications.
- A structured agent-status protocol carried by yet another terminal escape sequence, OSC 7501, then exposed through JSON-RPC.
"Nothing" says unfiltered, raw human intimacy quite like reading output from a pseudo-terminal file descriptor and watching \x1b[31m error logs cascade through an SSH pipeline and "pseudo" TTY device drivers.
Wrapping this tech in branding that sounds like a sultry, pseudo-erotic pitch is peak HBO Silicon Valley. TTY literally stands for teletype. We are supposedly building the frontier of autonomous artificial intelligence, yet our cutting-edge orchestration layer still routes interaction through abstractions inherited from noisy, electromechanical printing telegraphs of the 1960s. It's not a "native home for your displaced AI souls" -- it's an over-engineered wrapper shackled to the physical heritage of paper rolls and mechanical carriage-return timing: 110-baud Teletypes, separate carriage-return and line-feed operations, and terminal null padding conventions devised to give machinery time to move before the next character printed.
https://en.wikipedia.org/wiki/Teletype_Model_33
Didn’t know that about Brooke Shields, I’ll try to remove that phrase from my vernacular.
That said the rest of it is, well, complicated. Of course you can come up with something cool and new that doesn’t require all that plumbing, but how do you get people to use it?
If you build a beautiful piece of software that no-one uses, what was the point?
Cool report
I switched to herdr this year after a decade of using tmux, and I haven't looked back since. Herdr is wonderful, not only because of the coding agent integrations, but also because it uses libghostty under the hood. So no terminal-emulator-inside-another-terminal-emulator shenanigans.
If you prefer GUIs, I can recommend herdr-gpui which connects to a herdr daemon. It has a bunch of goodies like remote phone pairing, and a built-in browser integration to send screenshots and notes to coding agents.
https://github.com/penso/herdr-gpui
If you prefer GUIs, you can use the the new herdr desktop app :)
It will be good to get the remote feature in there though.
I've been using Zellij a long time and considered using Herdr, but the VC funding aspect scares me. I don't want to invest a bunch of time in a tool that's going to treat me as a product.
It's such a good experience. Try it out, later we fork if they pull anything funny.
I used Zellij in the past, it is also fantastic and intuitive product. I had issues with... I think selection and copying and pasting which made me abandon it as I was getting frustrated. Herdr has been wonderful and I've been using it for a while.
Pretty much the first thing I tried to set up with I heard about Zellij and decided to give it a shot a while back was how to scroll back and copy paste text without using the mouse, since that's one of the core features of tmux for me. If my recollection is correct, the closest thing that I was able to figure out was some sort of built-in way to open output in a vim buffer, with the info I found when googling basically saying that it was intentionally not something the developers felt like it was important to implement anything closer to tmux for. Maybe it's changed since then but I distinctly got the impression that it using it would force me to adapt to a bunch of new workflows that didn't resemble anything I had seen before, and it didn't really seem worth the effort.
I don't know, this seems reasonable. I already know how to select, copy, and manipulate text in my $EDITOR. Why would a terminal multiplexer try to implement all of that rather than just let me open the terminal in my $EDITOR?
I'm not opposed to using the same shortcuts as editors (which is what tmux already does; you can even pick between emacs and vim shortcuts, in addition to configuring your own); there's just a lot of other stuff I _don't_ want from a text editor for my existing terminal output, like, editing text.
I recently switched to Zellij with Ghostty and the selection/copy/pasting seems to work just fine. I previously used zellij and kitty and I would have issues.
Since I am using Ghostty I can give it a try, it would be nice to see it resolved as it was stupid tbh, all the difficult stuff is working well.
It's open source so if that changes you can always fork? I already have some personalizations accruing on my fork.
Same, but the cost of maintaining a fork fell through the floor in recent years. If a decent fork gains momentum, i can probably even stop spending tokens on mine.
But I’m already using ghostty. I installed herdr but made the mistake of launching it from within tmux and had no idea how to close it.
I then opened it somewhere else, but I had no idea how to interact with it. I also couldn’t get two other windows to “join” the same herdr session, so I gave up and went back to tmux.
For someone who has used screen / tmux for almost 30 years at this point, what does herdr do that tmux can’t?
I use Claude Code extensively. I also use plannotator extensively, which is how I found out about it.
It's all about agentic work. Tmux doesn't know what an Agent is. You use Pi, you use CC, you use Codex, ... they all have primary supports in Herdr with built-in integrations around each of these harnesses
But if you don't do any agentic work, Herdr has 0 value vs. tmux.
I use Claude Code extensively as I mentioned, but what does “primary support” look like? Why should tmux “know” what an agent is?
I just use tmux with multiple ghostty sessions and I have different Claude code sessions behind different tabs and panes and work like that.
What am I missing out on that justifies now using a normal terminal+tmux anymore?
lol at the other replies waffling at you without describing the things that makes herdr appealing:
* agentic inbox [see which agents are waiting for your input/working/idle etc] * agents can read, spawn, send-input etc other herdr panes * makes cross harness orchestration / communication easier
fwiw I liked herdr initially but bounced after like a week because ghostty + notifications + harness extensions gave me enough that it wasnt worth learning the new thing.
If tmux can know what an agent is, it can be leveraged to communicate with an instance of that agent (input and output) given it owns the PTY, and that terminal output can be structured somehow.
If it can communicate with the agent, it can be used to drive the agent - to send it tasks, and communicate the status of those tasks.
And so Herdr - and Superlogical, and others, is a tool which leverages PTY ownership and terminal protocols to allow orchestration of coding agents across vendors using their CLI surfaces.
But Claude Code and Codex and the rest of tui harnesses are also very good at just typing in each other's panes basically when I need to send a message over. It's been working for almost a year now. I have been more intrigued by T3 code or something similar, just to be able for harnesses to message across machines.
overall not so much, but basically if inside tmux you have 100 terminals opened and only 10 of those has 'agents' it will allow you switch quickly between those 10 terminals without spending too much time on locating those. also it will show statuses about agents, is it work done already and you may need to check it and stuff like this.
If you do Claude remote control plus Claude app/site you get all of that though...scratches head.
part of the point is so you don’t need to stick with claude/anthropic apps? you can use other models i’m assuming
I would assume other models have their own harnesses as well but I guess if you want it at all in one place that would make sense. I just need a single model.
This is more like a harness for harnesses imo? or a managing interface. some have their own
True, but tmux is very configurable
https://github.com/hiroppy/tmux-agent-sidebar
Has been pretty nice
How to properly secure it?
If you put it in a sandbox than herdr can no longer see it, can it?
The point about libghostty is that Herdr doesn't try to reimplement shell forwarding or emulation, instead it uses the same code your terminal already does. This avoids compatibility problems with more complex TUIs.
Running `herdr` always opens the same session, the app is a just a UI for a server running in the background. But there is no reason to open it in multiple windows, the point is to manage all shells from inside it.
Then you can from another machine do `herdr --remote xxx.local` (over Tailscale) to access the same session, or use `herdr machine add` to get your shells from all machines at once - I don't really use this part.
From someone who also used screen and tmux for 30 years: 3 reasons why I switched to herder. They may not be relevant to you and in that case tmux is just great. One: visual indication which agent is working and which is waiting for an input. Two: native support for remote herdr sessions - in the same view. Three: it’s responsive and can be completely touch-driven through menus on a small iOS screen in Termius.
I was looking into Herdr too, but could not bring myself to switch away from tmux yet (there are a couple of shortcuts that I have in my workflow, which I can't config in Herdr). What I ended up doing was to use the tmux agent sidebar, which, with mouse support, I think does a lot of the same things already.
All of this, plus it knows what agent is running in which windows and what session that agent is working on. If you have to reboot, it can reconstruct all your terminal setup and restart the agents on the right session automatically.
For me, the main reason is the remote sessions. I can copy and paste text and files in a remote session with no extra config, it's amazing!
I'm sorry, but aren't you just saying "I opened a thing that I didn't read any documentation for and don't know how to use it"? Did you know how to use tmux without reading the docs?
What's more, the sidebar literally has a thing that says "Menu". When you click it, one of the options is "detach". It's not rocket science...
Herdr is great, one of the exceptionally few TUI applications I actually use daily
On the note of libghostty it does get me very excited about the new multiplexer thing Hashimoto is building with it
Rex (by Hashimoto/Superlogical) is good. It’s really, really good. Feels and looks great.
This thing, presumably: https://www.superlogical.com/rex/docs/installation
> Herdr is wonderful, not only because of the coding agent integrations, but also because it uses libghostty under the hood. So no terminal-emulator-inside-another-terminal-emulator shenanigans.
Now I’m interested. Does it have basic feature parity? How does it compare to tmux in your experience?
I was able to have a language model configure herdr with the same bindings that I was using in tmux. It was a very easy transition from one to the other.
I started using it last week, the tui interface is very polished and it is mouse-friendly. It also allows you to configure the notification channels. Its a different beast from tmox, I'd say its quite more user-friendly and gives you a list of your running agents and their status directly on the interface.
I could not work without it anymore, such an easy solution for a common problem. Will definitely try out the app as well.
Really happy to see this!
herdr is by far the most polished tool in its category. Software geared towards agents is usually vibe coded -- and you can tell. There will be bugs and edge cases and it feels like no thought/care was put into it.
herdr doesn't feel that way to me. The marketing/landing page definitely has LLM-isms, but the application itself is very stable and well thought out.
A true UI is the next logical step and they seem to have done the obvious things. The one surprising bit that nobody seems to be talking about (yet) is the entire UI looks like its controlled via "recipes", so you can customize the UI in ways that you couldn't with traditional software: https://herdr.dev/layouts/
I hope this means an official mobile app is on the horizon!
> herdr is by far the most polished tool in its category.
Can't speak to this, but I like that I didn't feel like I needed to try an alternative. It just did exactly what I needed it to do.
That said - I couldn't find a way to move a pane to a new tab. I can swap panes within a single pane (select -> right click -> swap with focused pane), but I can't find a way to do it between tabs. Maybe it's a skill issue. Anyone have any ideas?
Although this isn't a direct answer to your question, but everything that can be done in Herdr is trivially discoverable by asking a CC (or equivalent) session inside Herdr. In fact a lot of the things I do in Herdr I do by asking an agent to do it for me, instead of trying to learn how to do it manually.
Some things might require a non-trivial workaround, but a coding agent will also happily build and test it for you.
Playing with this for a bit, the customization you can do is really cool. You can spin up an agent, tell it to add some functionality, and it all reloads instantly.
For example, I wanted to be able to hide chats from providers when my usage is out for the 5h/weekly period
I'm not sure how they're doing this considering it seems to be written in Rust
It is using JavaScript for user interface customization via:
https://gpui-kit.com/shell/
Agreed. I was really sceptical of it and was in the camp of "tmux does all I need" but it is honestly fantastic and ever since they added the ability to access multiple ssh machines from one herdr session I just live inside it.
Have you tried T3 Code? IMO it goes a step further on UI by dropping the idea of interacting via a terminal. Also it has had an official mobile app for months.
T3 Code, Paseo, Conductor, Orca, Jetbrains Air, Jetbrains Thinkrail, etc.
There's 30 million of these and they all look basically identical.
I haven’t tried T3 but I have tried Paseo and Conductor.
IMO it’s a great idea in theory but there are too many rough edges.
You probably haven’t tried Superlogical Rex then.
That's funny, I got tired of tmux the other day (it got into 100% spin loop from an unsigned int underflow used as a loop condition variable. and I debugged it all by myself like a big boy, without LLMs!) and built something that seems to be identical in architecture in terms of the "supervisor which allocates pty, just stream bytes, client decodes" part. I even use libghostty for the HW accelerated GUI I built. It's the right way to go imo and it's not too difficult to build... hence...
I don't really understand why they would turn this into a company. I become suspicious of products that raise a lot of VC money, implying that they expect a unicorn-ish valuation, where I don't see them being able to justify their valuation via obvious monetization techniques (e.g. selling a classic license for a GUI app where you get updates for some period of time, usually a year, and then you get to keep the software, but no longer get updates, just in case people forgot how things worked before subscriptions).
That usually implies something like selling cloud subscriptions, vendor lock-in, or selling data. I'm happy to be wrong, but it's just an extremely common pattern. And as an aside, I know there's an impulse to monetize everything, but I think we should probably fight that impulse.
Got my invite today. My jaw dropped.
People should put a screenshot of their desktop app in the README of their GitHub repos
agreed, this is a huge pet peeve of mine. every app that has a frontend should have screenshots on their github
Related:
Herdr Studio - https://news.ycombinator.com/item?id=49652188 - Sept 2026 (20 comments)
Connecting the machines - https://news.ycombinator.com/item?id=49612818 - Sept 2026 (35 comments)
Herdr now connects to multiple servers with a single client - https://news.ycombinator.com/item?id=49602323 - Sept 2026 (3 comments)
Herdr is joining Y Combinator. The runtime stays open - https://news.ycombinator.com/item?id=49201003 - Aug 2026 (189 comments)
Herdr: One terminal to rule them all - https://news.ycombinator.com/item?id=48756578 - July 2026 (178 comments)
Herdr: Agent multiplexer that lives in your terminal - https://news.ycombinator.com/item?id=48714802 - June 2026 (110 comments)
https://github.com/jkuepker/herdr-layouts/blob/main/lcars/pr...
This alone is worth trying the Desktop app
I'm sure I'll get bored once the novelty wears off
I'm already a happy user of herdr CLI (or whatever it is called. I use it from Ghostty)
I've been using Orca for this as it's the most popular on GitHub and seems to have more features. Is there anything notable that this offers over it?
Orca was too heavy and, ultimately, too slow for me too keep using. Switched to herdr haven't looked back.
Different philosophies. herdr has defined a few core primitives (machine -> space -> tab -> pane -> terminal) and lets you customize all the ui around it. Among all the ADEs I've used this is most reminiscent of emacs.
Orca is a bit heavy in asking you to define projects / worktrees up front. Anything you do, even quick edits starts with a worktree. Projects get their own modal and file browser before you start working. Herdr just lets you cd, mkdir, get going. But it's also worktree aware and organizes them under your repo when you do create them.
Herdr looks awesome, exactly what I need. However, automatic updates worry me in the light of recent supply chain attacks. I don't want be in a situation where one day I just find out I'm running malware on my computer. I'm not sure if there's an option to turn it off.
Darn, title had me thinking they released a mobile app to compliment the TUI daemon.
termux+mosh is fine, but especially on a phone I'd love a slightly more intuitive (or really, efficient) UI. Especially as mosh breaks touch/mouse input on the remote host.
I've never understood the point of the desktop GUI harnesses to run TUI (or at the very least, primarily text-based) programs anyway, but perhaps they're better for onboarding less tech-savvy users.
I've been using a Collie on Android to connect to my herdr server, it's very good
I use https://github.com/0cv/herdr-mobile-relay
I use it only with tailscale.
Have you tried moshi? It’s Herdr aware.
Tui-in-guis like orca are nice mostly because of sensible keybinds without prefix or leader
Moshi is pretty fantastic. I’ve been using it for several months and it works great. The developer is also launching fixes and improvements every week. It’s gotten materially better over the time I’ve been using it and it started out pretty great.
Looks good but not way I'm paying a subscription for a tool like this, make me pay a one time fee and give 1 year of updates (90€ would still be too much for me, 50€ I might reconsider).
You can already do that. They have a one-off payment forever
Ah sorry didn't open my eyes apparently, still a bit expensive to me but I will probably test drive it then, hope the desktop app is good it's probably even more important than the mobile app to me and does not seem like their focus at all.
I'm still not convinced about paying it. I'm wondering if something similar can be vibecoded for personal use
I'm using Ovis a plugin for herdr that adds a web app wich works well enough on web and desktop, but I feel like I'm playing Russian roulette every time I'm updating it and there's updates every day... It's the trust issue of vibe coded apps... Moshi I would feel slightly more confident they won't voluntarily add malware to it and pay attention to security.
If you are on iOS, Heeler is a great Herdr client.
try cosyra mobile. Its the best mobile client for herdr. It uses websocket and is has a really smooth ux.
This is a paid per month app? Why?
[flagged]
In this vein I’ve started using Tern (https://stencil.so/tern), but it’s closed beta, from the creator of Oh My Pi.
This space seems to be heating up, everyone wanting to build the single pane of glass for agentic work.
I got access this week and I am blown away at how much better it makes oh-my-pi, and omp was already great!
I got access around a week ago but found it quite unintuitive and found myself sticking to herdr for now, but was probably a bit quick to judge. What makes tern great for you?
I've got access weeks ago, seems I've completely misunderstood what it does!
There's still space for innovation and UIUX improvements so why not I say
Wow, that site is gorgeous
Having used herdr since its release, the recent 'Emacs is 50' made me decide to go back to it and just vibecode herdr features i use into an emacs package.
It works great so far. I can just ask it to update its layout whenever i think something better is possible. I wont be going back, but the one thing that might change that is a mobile app.
this new herdr app is actually very close to emacs in philosophy. Instead of the C core and lisp runtime herdr has GPUI (rust gpu accelerated UI drawing) and herdr-core which is the agent daemon. There's some vocabulary for spaces tabs panes agents, and js sits on top to define UI/UX. Very little is hard coded, and default app functionality is implemented in the same way that users can customize, and changes hotreload without restart. Almost everything seems to be just js/json just like most emacs is just elisp.
I'd love to give it a spin if you ever publish it.
Another vote for publishing it, would love something like this in Emacs.
This is the first time I have even hear of Herdr, and I try to make an effort to follow ai.
It's getting very difficult to keep up with all the agent harnesses coming out. I love ai but wow its hard to keep up.
herdr is moderately well known if you look into agent multiplexers. IMO it is the best one from what I have tried (Paseo and Conductor are competitors in this area)
It's not a harness -- it's really more like a tmux for agents.
I haven't used it yet, but isn't it also a Harness? Or do you run Pi or Claude CLI or whatever from inside Herdr?
No it’s not a harness at all!
A harness is what does the model loop, tool calls, etc
herdr does none of this. It’s really just a tmux/terminal geared towards organizing harnesses like Claude Code or Codex CLI
It’s a really great approach because you can use whatever cli you want, or even a mix.
[flagged]
Your comment is so comically dumb.. Herdr is yet another recently vibe coded ai helper in a niche category. You wouldn't miss anything about frontier ai or coding agents if you don't use herdr and/or terminal multiplexers in your workflow.
I have been using herdr for a few months. Then i bought a new machine, installed claude code and asked it to setup herdr. Shockingly claude said it has no idea what herdr was. This was weird.
So I wouldn’t blame the commenter above.
The Herdr website [1] has one of the best demos I have seen, one that I think would blow away those who aren’t nearly as AI-brained as I am. Why are we locked into old, boring UIs when I can just ask the agent for something like cows in a pasture or Windows 98?
[1] https://herdr.dev
We're not, plenty of applications support custom themes.
I have never used an application where I can type in anything close to "make this look like Windows 98" and it just works.
I mean one where you can say to herdr, make that look like Windows 98. Still can’t think of a good example that’s not already from Windows 98 unfortunately.
For all applications that support theming you can ask an agent "make this look like windows 98"
don't quite understand how typing it into a popup instead of an agent makes it revolutionary
News to me, I guess. I have never seen anything like these demos. Most themes in software I use change colors or rearrange existing UI elements rather than create kanban boards and floating interactive cows where they did not exist before.
Don’t get me wrong, I think having much more customizability now that maintaining complex configurations is easier is absolutely the direction we should be headed in. I don’t think floating cows are going to be the ultimate goal, but it is a step in that direction.
But this sort of theming specifically is something that was experimented with a couple decades ago and given up because people wanted simplicity and consistency instead. It’ll be exciting if LLMs reverse the trend.
I loved Herdr at first and got a massive productivity boost. Installed a bunch of plugins and created a few myself, but felt a bit messy after a while. When working on 10+ tasks I had a hard time getting a proper overview of the agents and high cognitive load when swapping between the tasks.
I built my own app based on what I liked with herdr and added some other things, but with this new herdr update I might switch back https://svall.dev/
Still a work-in-progress[1], but I have a mobile app for herdr also built with a native GPU interface (but not using gpui), optional LLM usage (outside of the agents themselves), designed for a hands off voice-enabled approach to development. The STT/TTS are both on-device. Yes, I used agents to build it, and you can see the development record in tree under docs/design.
[1] https://github.com/flankrapp/flankr
What’s the planned monetization model for all the terminal multiplexers backed by VC funding like Herdr, Rex, etc?
Being acquired for $10B by $1T companies.
Yes, become too big not to be acquired.
If the terminal is just a screen to a remote, they will sell you the remote, completely integrated into the harness.
I think they are all planning to make money on selling cloud or inference
> Usage telemetry is on by default in releases. Turn it off in Settings → Privacy.
Probably a good idea if you work with sensitive data (and you’re already on Zero Data Retention)
I love the "What we send" information - I looked that and figured "that's not too bad - fine, have it!". I think this is the first time I've seen this kind of transparency.
Telemetry does not include any data from your terminals, or names of files, folders or projects. In fact they even provide an example JSON payload for you to inspect in the settings dialog.
Supports SSH to remote machines out of the box, cool. I wonder if the protocol is more efficient than the TUI over SSH, would be nice on crappy connections when traveling.
the poor experience on a crappy network connection is an SSH problem unfortunately
It is, but if you send more efficient deltas, batch packets more intelligently, and introduce non-blocking UI tricks, the experience can be improved by quite a bit. mosh demonstrates some of that.
this is a solved problem with mosh
true, BUT mosh does have its tradeoffs. I actually prefer using tdp but using jwt for auth and websocket for transfer. Best of both worlds
If you like Herdr, you might also appreciate https://github.com/BrokkAi/mjolnir/, which is similar in its multiplexing features (although built on ACP) but offers cross-provider subagents, memory and session log sync, live migration of sessions across profiles and providers, and load balancing across multiple subscriptions.
Does anyone have experience in how herdr compares to Zed?
I'm currently using Zed as my main "agent orchestrator" and IDE because it handles various simultaneous agents quite well, using different harnesses. I'm especially a fan of the diffs for code review and the ability to easily highlight code and add it to context. The way it handles worktrees and git is also nice, IMO.
Can you use Zed with your regular Claude subscription?
Yes, if you have an active Claude subscription, you can connect to your existing installation using ACP (https://zed.dev/acp).
I have had much better experiences with the desktop apps for managing LLMs compared to the terminal.
Tasks like switching between multiple sessions, copying and pasting, viewing diagrams/web pages, and resuming remote sessions I have found significantly more polished when using the desktop app equivalents.
How does it compare to CMUX?
[flagged]
I don't think I need it, but Herdr has been absolutely rock solid for me. And I have this app Pairfob that I can check things up on the phone if I really want to. And while I don't use it quite as often, seems to be more solid then Chatgpt (Codex) and other apps.
I have 2 machines and laptop linked together and running things have never been easier.
I left tmux+vscode a few months ago and haven’t looked back. Clearly this new thing is better than flipping between a dozen tmux panes and other windows.
But I’m using my own “IDE” inspired by herdr and pane, worktrees down the left and panes on the right, but besides terminals the types of panes also include:
- A native Obsidian-level markdown editor with support for comments that agents can see - Claude code with chrome for context size, and a custom UI that looks more like the web - A diff viewer with support for comments - A browser - Can tag worktrees with PRs and issues - Everything (except the browser) works seamlessly over ssh
It’s all zig+gtk, all vibe coded. I rarely open vscode these days.
Maybe the herdr plugin ecosystem can support all this too. I kinda want to get out of the business of hand-polishing UI.
I actually did the same but at some point was like "this is good enough" and it works well. Diffs, comments on diffs, integration with Linear and worktrees is good enough for now and gives me a productivity boost. Couple of other apps in the space but at the time none supported Claude Code in terminals
I recently searched for an application to manage my agents and work streams, on top of Claude Code. Chat called them "task cockpits", but I'm sure I've seen other terms used for this category. I found a bunch that were abandoned already. And I didn't see Herdr until I already selected Pane, and committed to giving that one an honest try. It makes me wonder, is there any winners in this space yet, or is it just as fragmented as it seems?
I can’t seem to understand if this is something that could also be used without agents. I see the appeal of having first support for agents but I’d also like something I can use myself like I use tmux, without agents.
If agents didn’t exist, would herdr provide a nice UX as a terminal multiplexer?
Does herdr have a mobile app like Orca?
I think you are supposed to use Collie as your mobile app for herdr, but I'm still trying to figure this out, so I could be wrong
Installed Herdr's Desktop App. Opened new Terminal. Opens bash from /bin/bash instead of bash from Homebrew (/opt/homebrew/bin/bash)
Spent a while searching for how to change this.
Finally asked the claude session inside of Herdr how to do this.
Burnt through like 10% of my weekly limit with Pro writing a mod to do this.. which seems kind of like overkill
If you like the idea of herdr, but want something that isn't opinionated on terminal vs. desktop app and has mobile support, you might like https://github.com/codealong/codealong
Nice! I have been using herdr for a few months, love it and for me it is a step up from years using tmux. The new native Herdr app is yet another step up. I use multiple Macs with Tailscale connections and the Herdr app makes jumping around to different projects and different machines easier.
The relevant question: is herdr going to be sunset by herdr-app backed by herdr-core?
To me this is a valid signal.
I was wondering this, too. The desktop appears to be based on herdr-core, and naturally I'd have imagined Herdr Server (terminal) to migrate to herdr-core as well. In fact I'm not even sure why they built herdr-core when they had Herdr Server ..
Anyway, the best comment I could find is that one here where even the main dev doesn't know the direction .. https://github.com/herdrdev/herdr-app/issues/63#issuecomment...
Lack of vision and fragmentation is concerning
I was initially excited because I thought I could just try out a new frontend while keeping my 10 existing spaces each with 10 worktrees each with 2 or more tabs untouched.
Same here - just started to really get going with herdr not sure I want to worry about moving again
Haven't used Herdr before - was looking at README and was looking to check screenshots and documentation. However, Chrome reports the SSL cert to the site https://herdr.dev as invalid. Will wait till it gets resolved.
I think it's just you https://www.ssllabs.com/ssltest/analyze.html?d=herder.dev&la...
It works now. Haven't restarted browser/mac - so not sure what went wrong.
Anyways, Herdr looks powerful - will try it soon. One thing I find missing in every single one today, is that they do not allow using a different AI account based on the workspace/project you are on. It would come in handy in case of shared family machine.
I don’t think they need to “allow” this specifically. Every new pane opens in a shell prompt. You must start the agent from there, so you can start each agent with different keys/accounts already.
I understand this got funded by YC. But how is it supposed to make money?
herdr cloud
> The goal is to link your machines with a single command and stop worrying about how to reach them. Herdr Cloud would provide the middleman layer for those connections.
https://herdr.dev/blog/connecting-the-machines/
The examples of this new desktop app already use a hc command which could hint at that service (my guess)
How does it compare to Rex? (I use Rex but haven’t tried herdr so cant compare yet)
I've been waiting for Rex access! How are you using it, have you installed on different servers?
I have more faith in team behind Rex so not putting the time in for Herdr.
I've been using Rex for the past week, almost entirely through the macOS client and it's been great. Replaced ghostty+herdr (the TUI) for me.
I have its local server and an instance of the alpha linux server on my cloud based dev box, hooked up via my tailnet.
I've played around with the iOS client a bit, but haven't had the need to do terminal stuff from my phone.
Rex is cool, but I feel gimped because all the cool stuff happens over Tailscale but it doesn't support headscale.
Actually I haven’t used the tailscale stuff. I just ask my agent “to use rex” and it maintains my services and restarts them accordingly
Right now the new herdr has more ... just more. Rex has the polish, attention to detail, and attention to performance.
They both follow the same "gui with a cli to manage the engine which owns the PTYs, has some interaction protocol, wraps coding agent interfaces, and you can connect to remote machines" model.
Unfortunately they are both GUI only atm and so I'm still using herdr-tui over ssh until it breaks or we get a tui surface.
I love the current model, where I can use the tui in ghosttty or wezterm or iterm, or pop an ssh session, and if it makes sense there's a gui version for whatever device I'm using. It's like magic, I'm free to do whatever from wherever. Forcing a GUI feels like them asserting ownership or control. Now I need their mobile app if i want to develop remotely, etc. I recognize the irony, the core is open- I'm free to develop against it - but we all know that realistically only a handful will and those will lag. Also i am sure that GUI is first because that's where the users are. Eh.
Didn't have much time to try it this morning, but so far the Herdr app looks fantastic. I am already using herdr on the console to control Claude on my Mac Mini, but having to log in again after closing my Macbook was always annoying. And I missed seeing the code diffs and being able to commit changes in VSCode, which is still my preferred development environment today (with the Claude VSCode extension).
But Herdr app addresses both issues. I don't need to reconnect, and it has a diff tool that allows me to review and commit. Looking forward to work with it!
This new GUI looks awesome, however it's not clear how to do the agent integration? With the TUI I installed the skill via `npx skills add https://github.com/herdrdev/herdr --skill herdr` and the agent knew what I was talking about when I said herdr, now I have to tell the agent 'Read `herdr-app --skill` and ...' - seems like extra steps?
Is anyone else finding that it doesn't pick up Claude Code agents? My Agents pane stays empty no matter how many instances I start and work in. Appears to continue to work fine in herdr, though.
I just got started with https://www.onorca.dev and I like it, how is herdr.app different if anyone knows?
Uses less resources since it's mostly Rust, to my knowledge. Whereas Orca uses Electron.
I hope they release a mobile app too
Have you tried Moshi? Integrates with Herdr TUI
Not the new one though. I'm sure it'll come. The new herdr is not backwards compatible, and the TUI is unlikely to see development (at least in its current form).
Agree. I'm also not happy with how the TUI and the app are not using the same "server"
Me too in the meantime I'm using https://github.com/moezakura/mux-pod but the UI is terrible
[flagged]
A native app is definitely a good thing.
herdr is great, and the original runs-in-a-terminal vision was cool, but at least on OSX it has a ton of selection management bugs that are pretty sucky.
As I understood, this is a completely different thing from what we had so far (herdr TUI), fully incompatible - so technically the only common thing between them is the team and the name.
It doesn't see the sessions in my already-running remote herdr server, you cannot access the Herdr App remote sessions from your mobile phone (previously you could via SSH client like Termius).
I was actually kinda pissed when i realized this. Then i got an invite to Rex from superlogical, and I was free! Only to find out it's the same damn thing :)
Heeler has a PR open to target the new core, but no sign anyone is working on a tui for it :(
[flagged]
Which of these various new multiplexers (herdr, cmux, etc) work over ssh, and which have protocols underneath?
I've been scripting tmux for a while now, and it's not the greatest. And when I have a bunch of opencode sessions rolling along often my tmux could usage is smoking. I'd be happy to try other things. But I'd way rather get into something with protocols, that is open to play.
Hydra might be worth a look. The daemon owns the various agent processes (any that speaks ACP), and everything else is a client of open interfaces: a REST API for managing sessions, and ACP over a WebSocket for attaching to a live one. The TUI, a browser UI and VS Code all use those. So you can script it with plain HTTP instead of tmux. It doesn't use ssh; remote access is TLS. The TUI shows one session at a time, and ^P jumps between them.
https://github.com/smagnuso/hydra-acp
herdr has native support for remote machines. Both the CLI and the new app have this.
At work I do much of my work on a ec2 box via ssh with herdr
Rex will as well apparently
Not having a normal installer/uninstaller is a weird trend, I dislike these powershell scripts that dump stuff all over and have no way to uninstall it after.
Especially when it's an AI agent tool and an AI agent can probably make an EXE installer in a few minutes.
I just had a look at the site and screenshots, I usually run my coding agents containerized (one per project), anyone else doing this and wiring it to herdr? is it possible? from a quick glance it seems to me it installs cc/codex/opencode/etc on it's own and just does multiplexing
I have a project that auto-containerizes with its own frontend, pretty sure you could wire the backend to herdr but haven’t done it yet: https://yaac.dev/
I'm working on https://corb.ax from the containerization side. I can check on integrating it with Herdr.
kinda related, anyone using simpler workflow orchestrators like dagu? (https://github.com/dagucloud/dagu I am not affiliated)
at the end of a day I started to one-shot some simpler workflows; e.g. I already have various scripts for spinning work-trees, setting up deps, picking up a clearly defined issue, waiting for some delayed CI checks, doing reviews with different agents, workflow waiting for another workflow etc. - without passing context between
Any resource detailing how you do this? I've been looking at dagu specifically for this.
depends on what you're looking for - for whole worktrees setup I use worktrunk [1] which handles all stuff like spinning up deps, containers, envs so ports do not clash, every worktree have clean slate so local tests work etc. solves like _a lot_ of problems I ran with worktrees in real-life apps
I started using dagu fairly recently so the rest is kinda a ball of various ruby scripts at this point so I can't give any good pointers here as I have no idea yet xd but overall I have some deterministic stuff (pick up some issue, setup worktree first, resume if crashed, implement, review and so on) build around just running claude/codex (one-shot)
there are _tons_ of solution for orchestration so it's really hard to choose from tbh; I gave herdr try this morning and ported my scripts so it works with their desktop app - looks kinda nice I guess; dagu have scheduler as well which is cool, but connecting different machines to single herdr server also looks kinda fun... so I guess I will continue experimenting until then ;)
[1] https://worktrunk.dev/
edit: damn scripting herdr with mods open new window of opportunities, I dig it!
Looks good but doesn't have a browser yes, like all the desktop applications of major harnesses have. Having a browser accessible to both the user and the model is extremely useful. I know you can do it via MCP but still, having a baked in one is better.
It's hard to beat Claude Code with /remote. I can easily keep things moving from my laptop, phone, or another desktop. I realize Herdr can technically do that too and across more agents, but it's more to setup and I really don't need that.
I've been building something similar except it uses iroh for P2P instead of SSH. Would really like to use this but I don't want to use SSH.
edit: nevermind I see it presents a console rather than wrapping the agent with a native GUI. This won't work well on high latency connections.
I also built a similar thing a many. And in my experiments (even though iroh is great) iroh took significantly longer to reconnect vs mosh and ethernal. And actually plain old autossh was the best XP for me to close laptop get on traincopen again on mobile Internet.
So eventually I removed everything other than autossh.
Seeing as how many people are building similar things I'd love to share more. Maybe wanna ad each other respectively to the repo's ?I can share some of my benchmark results
For me connect time is not really a concern, it happens once when I open the app. In my project I'm wrapping agents using ACP and presenting an actual GUI, so latency doesn't really have a perceptible impact. Running Claude Code in a console over any kind of remote connection is gonna feel bad.
Rather than for network performance I'm using iroh because I don't want to need a direct network path to the machines running my agents. Yeah I have Tailscale but that's more work to set up and I don't want my app to depend on users have a TS network.
To elaborate on my usecase that herdr didn't so at all the the beginning but still does not do well: I have many vm's in different clouds working on different projects with different secrets that I want to keep isolated ( now also at home lan machines which when im travel get aliased to their tail scale address)
So my biggest requirements are around making isolation between environments convenient but safe from supply chain attacks .
I wonder what usecase you were optimizing for
This sounds cool and a great use case for Iroh
while they do this, i hope they do not drop support for regular/other terminals
ignoring ALL the AI/agent features, herdr has been a decent replacement for tmux with saner defaults
so much so that i'm reaching for installing herdr instead of tmux and i would like to not be locked to a "desktop app" for it
+1 I tried quite hard to make things work with tmux, zellij, and even some light custom stuff, but if you have multiple agents running on multiple boxes herdr just works unreasonably well. I really didn't want to like what seemed like a random mix of features, but they've stabilized it to a sensible capability set.
Glad to see the desktop app and after did some research seems its like rex and with some once feature and it even have a Win98 easter egg if you can find it:)
Currently I am not switch to it because I just have my own app talk to the herdr daemon via my apps builtin tailscale and it works great
Just finished ricing herdr-app with pi, damn it looks good!
https://github.com/kyle-avrett/herdr-square-dracula
The app doesn't even work as advertised. Opened the app, started a new session, closed the app, and my session was gone. I thought this was supposed to be persistent sessions. I think i'll keep using Cmux.
Hmmm I tried it, but it didn't detect my open herdr sessions in the added machine. I don't see much benefit over herdr in Terminal/Ghostty, and I need to reconfigure my agents to make it work too.
There is a Windows 98 mode w/ minesweeper, solitaire, and clippy!
Edit: oh neat! Its self editing.
Anyone using this or similar with self-hosted models/engines (not on local machine)? Didn't immediately see anything about that anywhere in the documentation.
Fantastic news, I may have to switch back (or combine this with Paseo on my phone). I hope the RAM usage is a lot better as Paseo tends to OOM now and again (the desktop app).
The tabs mechanism didn't do it for me so I rolled my out (self-plug at [0]).
While it's not "agent-first" thanks to rich MCP I often use direct terminal control from one orchestrator agent to drive another one, and I haven't seen any generic solution like that.
Since I'm moving my work VM yesterday I even started working on ssh persistence using planted cross-compiled PTY host library. If anyone's interested check it out, I expect to roll it out within a week or so.
[0]: https://race-term.com
I've done a little poking around and there is one glaring annoyance. Why didn't they keep the keymap from the tui?
This is just so cool. I've been spending all morning having my agents mod and tweak herdr to fit my workflow exactly
Can someone tell me if I should invest in any of these or just wait until mitchel's multiplexer is out?
But can I install herdr desktop app and connect to my SSH sessions? I'm not running herdr on desktops usually
I should have tried before asking
How does it compare to cmux and wave terminal?
It works great in Ghostty not sure if I need another standalone app. I’ll definitely try it out but my expectations are low.
I prefer the TUI in ghostty and even that’s just a stopgap until superlogical is out of alpha.
In its current incarnation, superlogical is exactly the same as the new herdr, with the same gaps. A TUI does not appear in their backlog either. They are a native app.
Yes, but Rex is built on libghostty with the multiplexing and agent integration built in, built by people I trust who don’t let the LLM write all their docs.
Yep it’s awesome. I’m going to integrate the core engine to my mobile app so I can control it remotely.
Collie already does pretty well. https://github.com/AltanS/collie
Just termius + tmux works for me on my Android phone and iPad. I have a ntfy topic to alert from Pi when it needs something.
[flagged]
have you tried cosyra or moshi before?
From my limited understanding it's fancy tmux?
Please correct me if I'm wrong.
Please lmk what the advantages are!
Am I missing something, or is there no way to use plugins with the herdr app?
I've been holding this at arms length. Interested, but I'm so stuck on tmux.
the fork from the original herdr tui (without any explanation of the plan or reasons) is quite confusing given that they are using different backends
Looks like someone is going after Rex
Is herdr in the similar park as superset?
My cat decided to sit on the power button of my laptop, and upon restart herdr-app greeted me with empty terminal panes, whereas herdr-tui was just fine.
And seems like they use "herdr-core" (hc) to actually run sessions instead of herdr server, that flushes events only every 100ms -- the agent output is quite jittery with that.
Meh :/
I use herdr daily. One thing I’ve always missed from all of these solutions, though, is the ability to actually make code changes..none of them seem to offer that properly. Am I the only one? Is the goal to abandon code entirely?
Another desktop application without a screenshot in the Readme.. sigh..
Their website has lots of screenshots
feels like they're trying to get ahead of superlogical's rex.
took all of installing it on windows and setting up my mac as an SSH machine to see the slop all through the UI. i'll just wait for rex's windows client i guess.
[dead]
I really enjoyed herdr at first but ever since they got VC funding I am avoiding it. Enshittification and cloud features due to vc/monetization pressure incoming.
Since you used herdr before it got funding, were you going to donate or pay for it without VC funding?
I guess there is some nuance here. Would I pay for a very good agent multiplexer that makes me a lot more productive? Sure, same reason I pay for other apps or services.
But for Herdr specifically? Probably not. There are tons of similar free alternatives and even vibe coding your own is very easy with all the open source code of both herdr and all the alternatives available.
> Would I pay for a very good agent multiplexer that makes me a lot more productive?
That is not the question I am asking.
> But for Herdr specifically? Probably not.
So no then.
That is why they went for VC money since the author knew that donations weren't enough to maintain it in the open.
> There are tons of similar free alternatives and even vibe coding your own is very easy with all the open source code of both herdr and all the alternatives available.
Why would I waste tokens vibe coding something that would go unmaintained in 6 months (because it is "free") over a properly funded project that will last much longer than that?
The cost of maintenance is never "free".
You dont have to vibecode and maintain it when there are 20 other projects that do the same or similar thing. And maintenance is not free, but you know waht's even more annoying? Getting locked into a product that is pressured to achieve VC returns and thus becomes enshittified. There is no moat here and it's painfully obvious that open source will win in this category in the long term.
> Getting locked into a product that is pressured to achieve VC returns and thus becomes enshittified. There is no moat here and it's painfully obvious that open source will win in this category in the long term.
Herdr is...open source? and is also compatible with the free software principles. So where is the "lock in"?
You have not answered how will Herdr (and others) fund themselves to maintain the project without some form of sustainable funding without VC funding?
Not that I support VC funding per se, but without meaningful donations to support the maintainers, this is usually what they end up doing.
Are you asking the same question to emacs users?
Yes. So you donated to Emacs as a long time user?
No, and i don't feel bad about it.
Good. So as you as a long time Emacs user, no donations to support them at all.
Thank you for proving my point.
Don't confuse Open Source with an enshittification backdoor with Free Software.
Nope. I think you are confused since Herdr was free software and licensed under AGPL 3.0 [0] before it was VC funded.
One (emacs) gets millions in donations via the FSF and the other get little to no money or donations for being "free software" from the start despite both of them being "free software".
[0] https://github.com/herdrdev/herdr/blob/a57b97285c3308fbdcc16...
is this similar to zmx?
I love herdr, nice work team!
u can do all this with react and tailwind tho lol.
2004 called. They want to return their disemvoweled penultimate "e".
2004 - The Big Bang (Flickr) - Co-founder Caterina Fake wants to buy Flicker.com, but an offline beer brewer refuses to sell it. Out of options, she suggests dropping the "e". It works, and Flickr is born, instantly setting a precedent that typos equal modernism.
2005–2006 - The YC Copycat Flood - Paul Graham launches Y Combinator, and the application form might as well have had a checkbox for "did you delete a vowel to look disruptive?" Scribd (YC S06) drops the E because buying a real dictionary noun requires actual revenue. Twitpic and Pixlr emerge to capture that sweet, sweet pre-seed funding. Even Twitter originally launches in 2006 as "Twttr" because the actual domain is owned by a bird enthusiast. (They buy the "e" back six months later once they realize they look ridiculous).
2007 - The Peak of Style - Tumblr launches, proving the vowel-less trend isn't a passing fad but the official language of the new internet. Dropping the "e" becomes shorthand for "we have venture capital, a minimalist mobile interface, and absolutely no monetization strategy."
2009 - The Aggressive Pivot - Grindr launches. The convention branches out of pure utility and enters cultural lifestyle apps, cementing "-r" as the universal suffix for a location-based mobile service. Every YC batch now features at least three "Uber for X" startups named after a mangled verb.
2011–2012 - The Saturation Point - Companies like Blendr and Fiverr continue the trend, but the public begins to push back. Every SaaS tool on Hacker News looks like a typo. Critics and users start pointing out how incredibly annoying it is to vocally explain a business name ("It's called Taskr... no, without the E").
2015–2018 - The Death Kneel & Corrections - The joke gets old. Brand consultants begin labeling creative misspelling a "cardinal sin". Startups realize they are losing millions of organic visitors to the properly-spelled domains owned by domain squatters. The internet enters the "-ify" (Spotify, Shopify) and "-ly" (Bitly) eras, completely abandoning the broken suffix because pg wrote an essay saying missing vowels hurt your SEO.
2019–Present - The "Adult in the Room" Era - The trend is officially dead. Modern YC startups pivot toward ultra-literal names (The Browser Company) or aggressive, serious abstract words (OpenAI, CoreWeave) to pretend they are enterprise-ready AI infrastructure rather than a wrapper wrapper wrapper.
Postscript - The Ultimate Regression - The convention reaches its absolute, braindead rock bottom with Elon Musk. After burning 44 billion dollars to buy Twitter just so he could unban literal Nazis and amplify white supremacist conspiracy theories to the front page, he rebranded the entire platform to "X.com". It turns out "X" is just "Ex" without the penultimate "e" -- a fittingly hollow, single-letter monument to a sociopathic tech bro who couldn't innovate a functional name if his taxpayer subsidies depended on it.
[flagged]
[dead]
[flagged]
[flagged]
[flagged]
[dead]
[dead]
[dead]
[flagged]
Herdr is truly awesome. Game changing almost. If you think of all the overly complex agent setups[1] and overly complex software factories[2] to write code... herdr and a pick-your-coding-agent replaces all that with a very simple one thing which just works.
There are many ways to do what herdr does, like tmux, and rex, but in a nutshell:
You launch cli things (claude code, cli tools etc) anywhere inside a herdr process. On your own laptop you magically stay connected to all of them and can monitor, send commands etc. And those processes can talk to each other and control each other too if u want!
For a single person with herdr+tailscale, you can make your own evolving software thingamajig across all the possible computers u have access to (pls don't call it a factory, we didn't call your fingers a factory in the dinosaur days did we).
[1]: https://www.8090.ai/ [2]: https://steve-yegge.medium.com/welcome-to-gas-town-4f25ee16d...
p.s if someone from cloudflare reads this, would love to have a herdr process in a worker so it appears on my laptop as a thing i can do stuff with. while idle it goes dormant, i send a keystroke and it wakes up, stays awake for a bit.
p.p.s herdr team if u read this can you have a built in so i can instantly launch a vm in one of the many provide a instant vm (or sandbox or whatever the cool kids call it these days). I have been using the railway TS sdk to quickly spin up and throw away sandboxes, but why not just integrate <choose-your-sandbox-provider> in herdr and that is one way to charge money - pay herdr or provide my API key for vm/sandbox provider. Bonus points if it can sleep and charge less. Sort of like fly's sprites but more awesome.
Can you explain more about why you need to distribute agents between computers? Why not just run them on your local computer?
Because my job has multiple servers with different needs and tooling, and I must do my development within the appropriate environment.
I'd also _much_ rather do my development on a persistent, always-on host so I can switch what computer I'm using for a head or turn off my computer and drive home while Claude works overnight.
It depends on use cases.
I have multiple projects running in different locations/timezones - so I don't often need to have an agent running on a pc elsewhere - but just being able to smartly use herdr (also using rex) to set up a persistent service on those computers removes all the friction which used to exist with ssh keys and other things.
So I could do all these things before, tools like herdr make it easier, smoother.
Now tmux and ssh keys enabled all this before. So in a sense, nothing new here. but its a bit like dropbox - i used rsync back then, dropbox just made it easier to setup sharing and use without thinking.
My current anecdotal example: if you're building cross-platform native apps, you need to run agents on macOS and Windows, because xcodebuild only runs on macOS and the Windows SDK is only available on Windows.
Because they use resources (tests, etc) and once you have a few running it impacts you.
sandboxing