terminal-native runtime for coding agents. built by @lumendriada.

Joined May 2026
7 Photos and videos
Pinned Tweet
new herdr update! agent session restore is now enabled by default for supported integrations. after a herdr restart or update, supported agents can automatically resume their native sessions right where they left off. also introducing release channels. a small thread:
13
8
135
14,996
herdr retweeted
this goes beyond hooks, so wrote something: agents are becoming runtimes. we need to draw a line: the model is the lab's product. the runtime in your terminal is not. developers should own their runtimes, and runtimes need standardization 👇 herdr.dev/blog/coding-agents…
for the last couple of days, i've been working on a refactor of herdr's agent detection system. all agents these days have hooks/events, so this should be super easy... right? a big no. only a couple of these agents support the full lifecycle; the rest are half-baked. @badlogicgames with pi and @thdxr with opencode did such an amazing job with their agents that working with them brings me tears of joy.
2
5
24
2,145
Jun 10
oh and one more thing that didn't make the release notes... something sneaky landed in the docs 👀 herdr.dev/docs/windows-beta
Jun 10
new herdr update is out: v0.6.9! a lot of fixes and additions, but this update's theme is agent stability. the way herdr understands what your agents are doing got a complete overhaul: two detection classes, hot-reloadable rules, and a debugger for when detection is wrong.
5
2
34
4,791
Jun 10
new herdr update is out: v0.6.9! a lot of fixes and additions, but this update's theme is agent stability. the way herdr understands what your agents are doing got a complete overhaul: two detection classes, hot-reloadable rules, and a debugger for when detection is wrong.
2
1
52
7,872
Jun 10
detection rules are no longer baked into the binary. they're hot-reloadable manifests! when an agent ships a ui tweak that breaks detection, the fix lands remotely and your running server picks it up. no herdr update, no restart.
1
5
579
Jun 10
also in the update: native session restore for kimi, droid, kilo and cursor, directional pane swap, pane zoom, toast controls, ~20 fixes. lastly, shoutout to @KimiDevs team. kimi 0.14.0 shipped the missing hooks, so kimi now reports full lifecycle state including interrupts. full notes: github.com/ogulcancelik/herd…
3
1
16
4,330
herdr now supports Fable 5! ok it doesn't 'support' anything, it's your terminal, you can run whatever you want. but everyone was posting and I didn't want to be left out :/
5
2
143
5,422
reject gui wrappers embrace traditional tui with improvements 🐏
TUIs can be easy! look at what right-click does in @herdrdev refreshing to see something that works with both the keyboard and the mouse. and all this would not have been possible without @ratatui_rs
3
2
59
4,750
herdr retweeted
if a mobile term has no native integration with @herdrdev, its not an AI native mobile term everyone using tmux now will eventually admit herdr is a better replacement let's see how this assertion goes
3
1
22
2,405
thanks @RichOBray for the awesome coverage of herdr on better stack!
Imagine tmux built for AI coding agents. Herdr runs Claude Code, Codex, and OpenCode in persistent panes with live status, keyboard nav, and SSH remote. Built in Rust with zero Electron. Could this replace your AI agent workflow?
4
1
24
2,268
herdr retweeted
moshi moshi 😽 3.3.0 is coming. two things to watch: - the workspace selector now shows all your herdr workspaces when there's only one active session. - and select mode switching — native selection steps aside for TUI. no more scrollbar to copy. 👇 there's more
3.3.0 is coming. quite some cool features for @herdrdev users
6
3
40
10,770
herdr retweeted
Please, stop running your agents in Zellij. I just found out about Herdr and terminal multiplexing ACTUALLY makes sense now.
Please, stop running your agents in tmux. I just found out about Zellij and terminal multiplexing finally makes sense.
44
3
357
54,412
new herdr update! agent session restore is now enabled by default for supported integrations. after a herdr restart or update, supported agents can automatically resume their native sessions right where they left off. also introducing release channels. a small thread:
13
8
135
14,996
herdr retweeted
.@herdrdev is cool. I am tired of doing back and forth with github in the browser, so I created my own clickable PR/issue viewer, inspired by gh-dash put that in the left pane, codex on the right. saves me so much time
1
2
21
2,214
herdr retweeted
herdr is a tmux-like agent multiplexer that lives in your terminal. It can split panes, has mouse support, is agent aware, detach/attach sessions, works over SSH, supports sound notifications and more. @lumendriada made herdr using @ratatui_rs and is Terminal Tool of the Week! ⭐️
2
6
64
4,924
as herdr grows, i'm slowing down stable releases a bit. starting today, stable releases will happen weekly or bi-weekly unless there's a hotfix. for many people, herdr is becoming part of their daily workflow. stability matters.
1
16
747
if you prefer living on the edge: herdr set channel preview preview builds receive new features and fixes before they land in stable! see docs for more: herdr.dev/docs/preview/insta…
6
709
herdr retweeted
herdr.dev is a fantastic piece of software. I was in the early stages of building something similar, but happy to throw it away cuz this is *so* much better.
19
21
455
53,367