Herdr Dev Stack
AdvancedKeep Claude Code, Codex, and other coding agents running in Herdr sessions you can reattach from any machine.
Published 27 September 2026
About Herdr Dev Stack
Herdr is an open-source terminal multiplexer built for coding agents. It runs as a lightweight background server that holds real terminal sessions open, so an agent keeps working after the terminal window closes or the network drops, and you reattach later from any machine. Each pane shows whether its agent is working, blocked, idle, or done, so a screen full of agents shows at a glance which one needs you. It ships as a single Rust binary for macOS, Linux, and Windows, free under the Apache 2.0 license.
In this stack agents and models are separate choices, and both allow more than one pick. Claude Code is the default agent, with Codex, OpenCode, Pi, Cursor CLI, GitHub Copilot CLI, and Grok Build alongside it, out of more than 20 agent CLIs Herdr recognizes. Each agent has a home model family, but the model layer isn't welded to the agent: OpenCode and Pi run almost any provider, Copilot CLI and Cursor CLI offer a model menu, and Codex can be pointed at other providers too.
Herdr manages sessions, not code. It doesn't create git worktrees or show diffs, so parallel agents on the same repository need their own worktrees or clones, which some agent CLIs create for themselves. A local socket API lets scripts and agents create panes, read each other's output, and react to state changes, which is how agent-to-agent handoffs get built on top of it.
On a laptop, Herdr keeps sessions alive when a terminal window closes, but its agents stop when the laptop sleeps. Herdr is designed with a server in mind for that reason: run it on a VPS or home server and remote mode attaches from your laptop over SSH with the full interface, panes and status included, while the agents keep working with the laptop shut. A phone reaches the same sessions through any SSH client, since Herdr has no mobile app or hosted service. The terminal emulator you run Herdr in is another common addition.
Key Features
- ✓Persistent agent sessions that survive closed terminals, restarts, and dropped connections
- ✓Live per-pane status (working, blocked, idle, done) across every running agent
- ✓Pick one or more agents: Claude Code, Codex, OpenCode, Pi, Cursor CLI, GitHub Copilot CLI, and Grok Build
- ✓Pick one or more models, from Claude, OpenAI, and Grok to Gemini, DeepSeek, and Kimi through OpenCode and Pi
- ✓Reattach from any machine, including over SSH, with local and remote workspaces side by side
- ✓Local socket API and plugin hooks for scripting and agent-to-agent coordination
- ✓Single dependency-free Rust binary for macOS, Linux, and Windows
When to Use Herdr Dev Stack
- →Running long agent tasks on a server that keep going after you close your laptop
- →Supervising several agents from one keyboard-driven screen and jumping to whichever is blocked
- →Checking in on the same agent sessions from a desktop, a laptop, and over SSH
- →Building scripted agent pipelines where one agent hands work to the next through the socket API
- →Developers who already live in tmux and want the same workflow with agent awareness added
Pros
- Free and Apache 2.0-licensed, a single lightweight binary with no Electron overhead
- Sessions survive disconnects, so agent progress is never lost to a closed window or dropped SSH link
- Works with agent CLIs unmodified, so each agent behaves exactly as it does on its own
- Socket API and plugins make it a base for custom multi-agent workflows, not just a pane manager
Cons
- Terminal-only, with no graphical diff review or pull-request panel
- No built-in worktree management, so keeping parallel agents out of each other's files is up to you
- Every agent needs its own subscription or API key, so costs add up across providers
- No mobile app; checking in from a phone means an SSH client on a small screen
Coding Agent Options for Herdr Dev Stack
Herdr Dev Stack with Claude Code
Claude Code is the default agent here: Anthropic's CLI, strong at multi-file changes and a natural pick for the hardest task in the session. It bills through a Claude plan from $20/month or per token via the API. It's built for Claude models; DeepSeek and GLM publish Anthropic-compatible endpoints it can point at, outside Anthropic's support.
Herdr Dev Stack with Codex
OpenAI's agent CLI, included with every ChatGPT plan down to the free tier. Codex in the pane next to Claude Code is the simplest way to compare two vendors on the same task. It defaults to OpenAI's models but officially supports other providers through its config, including OpenRouter and local models.
Herdr Dev Stack with OpenCode
Open-source and model-agnostic, OpenCode lets one pane run Gemini, DeepSeek, Kimi, a local model, or nearly any other provider with an API key, billed per token by that provider. The usual pick for long-running bulk work on a server, where a cheap model keeps costs down.
Herdr Dev Stack with Pi
Pi is a deliberately minimal, model-agnostic agent harness shaped through TypeScript extensions, with any provider's key behind it. It suits Herdr users who script their setup, since a lean agent plus Herdr's socket API makes custom multi-agent handoffs straightforward.
Herdr Dev Stack with Cursor CLI
Cursor's agent running in a terminal pane, for developers already on a Cursor plan ($20/month and up). It brings Cursor's model menu, Grok by default with Claude, GPT, and Gemini available, into Herdr without another subscription.
Herdr Dev Stack with GitHub Copilot CLI
GitHub's Copilot CLI, a good fit when a Copilot plan is already paid for: one subscription from $10/month covers a menu of Claude, GPT, and Gemini models, switched per session and metered in GitHub AI credits.
Herdr Dev Stack with Grok Build
xAI's coding CLI, which splits a task across its own sub-agents in separate worktrees. Give it its own pane for a third vendor's take on a hard task; it's built around xAI's Grok models and bills through an xAI plan or API key.
These are highlighted picks. To see all the tools, check the AI Coding Agents category.
LLM Options for Herdr Dev Stack
Herdr Dev Stack powered by Claude
Claude is the default model, the one Claude Code is built for, and a common choice for the pane doing the most careful reasoning. It's also selectable in OpenCode, Pi, Copilot CLI, and Cursor CLI, so one Claude key or plan can back more than one pane at once.
Herdr Dev Stack powered by OpenAI
OpenAI's GPT models are Codex's home family and are also reachable through OpenCode, Pi, Copilot CLI, and Cursor CLI. Picking them alongside Claude gives a second vendor's approach to the same problem, and a ChatGPT plan covers Codex usage without per-token billing.
Herdr Dev Stack powered by Google Gemini
Gemini's long context window suits repository-wide questions and large refactors, and its Flash models are fast and cheap for routine work. Reach it through OpenCode, Pi, Copilot CLI, or Cursor CLI, or through Google's own Gemini CLI in a pane of its own.
Herdr Dev Stack powered by DeepSeek
An open-weight model priced far below the premium providers per token, a good match for agents left running overnight on high-volume work like test generation or mechanical refactors. It runs through OpenCode or Pi, or behind Claude Code via DeepSeek's Anthropic-compatible endpoint.
Herdr Dev Stack powered by Kimi
Moonshot's open-weight Kimi models are another budget option tuned for agentic coding, with long context for large repositories. Use them through OpenCode or Pi, or through Kimi's own CLI in a Herdr pane.
Herdr Dev Stack powered by Grok
xAI's Grok models, the family Grok Build is built around and Cursor's default. Add Grok when a Grok Build or Cursor CLI pane is in the mix, or reach it through OpenCode or Pi with an xAI API key.
These are highlighted picks. To see all the tools, check the LLM category.
Version Control Options for Herdr Dev Stack
Herdr Dev Stack with GitHub
GitHub is the default here, with the largest pull-request ecosystem and the review bots and CI tools most agent workflows assume. Herdr itself works the same with any git host.
Herdr Dev Stack with GitLab
GitLab fits when source control has to be self-hosted or stay on your own infrastructure. Herdr only sees git, so agents push to GitLab exactly as they would to GitHub.
Herdr Dev Stack Add-ons
Each addition below extends this stack with a capability the base stack works fine without. None are required: include the ones your product actually needs when building this stack, and skip the rest.
Server Add-ons
Add a server when you want the agent to keep running somewhere other than your own laptop, reachable at any time and not tied to your machine staying on. Some agents also offer their own managed cloud sessions as an alternative to self-hosting; check the stack's own description for details.
Herdr Dev Stack with Hetzner
Hetzner is the default here: a 4 GB server, Claude Code's minimum for one agent, is about €5.49/mo, and 8 GB or more suits several panes of agents for little extra. Herdr itself is a small Rust binary, so the agents set the size, and data centers in Germany, Finland, the US, and Singapore keep SSH latency low wherever you are.
Herdr Dev Stack with DigitalOcean
DigitalOcean's 4 GB Droplet is $24/mo and 8 GB $48/mo, more per gigabyte than Hetzner, with more regions and one-click snapshots and backups. Its 1-Click Codex CLI, OpenCode, and Grok Build Droplets arrive with those agents installed, ready for Herdr panes; the $4 plan's 512 MB can't run one.
Herdr Dev Stack with Hostinger
Hostinger's KVM 1 has 4 GB of RAM at $6.49/mo introductory, renewing at $11.99/mo, enough for Herdr and one agent, and KVM 2 doubles that for more panes. Its Claude Code VPS template preinstalls Herdr's default agent, so only the Herdr binary is left to add.
Herdr Dev Stack with Raspberry Pi
A Raspberry Pi 5 with 4–16 GB of RAM is a one-time hardware cost instead of a monthly bill, and Herdr's lightweight binary is a natural fit for it. It suits a home server reached over Tailscale; memory caps the number of agents, with 8 GB fitting two or three and 16 GB a few more.
These are highlighted picks. To see all the tools, check the Hosting & Cloud category.
Remote Access Add-ons
Add remote access when you want to reach an agent running on another machine — a VPS or a home server — without exposing it to anyone but you.
Herdr Dev Stack with Tailscale
Add Tailscale when Herdr runs on a server and you want to attach to it from your laptop or another machine without exposing SSH to the public internet. The personal tier is free.
Terminal Add-ons
Add a terminal when you want a faster, more configurable place to run the agent than your OS default — most agent CLIs live here all day.
Herdr Dev Stack with Ghostty
A GPU-accelerated terminal with fast startup and native platform integration, a solid default for keeping Herdr open all day with many agent panes redrawing at once.
Herdr Dev Stack with Alacritty
A minimal, GPU-accelerated terminal focused purely on speed. It pairs well with Herdr, since Herdr already covers the splits and tabs Alacritty leaves out.
Herdr Dev Stack with WezTerm
A GPU-accelerated terminal with a Lua config and its own multiplexing. Its SSH domains are handy for jumping between a local Herdr and one running on a server.
Herdr Dev Stack with Kitty
A GPU-accelerated terminal with its own graphics protocol and scripting API, for developers who want deep customization around their Herdr session.
Herdr Dev Stack with iTerm2
A free, open-source macOS terminal with more than a decade of features, including profiles, triggers, and tmux integration. A comfortable home for Herdr on a Mac, though Herdr already covers the panes iTerm2 would otherwise split.
Code Review Add-ons
Add code review when you want an AI reading every pull request before it merges: it comments on the diff so bugs, security issues, and inconsistencies surface before a human has to catch them.
Herdr Dev Stack with CodeRabbit
Add CodeRabbit when several agents are opening pull requests faster than one person can read them: it reviews every GitHub or GitLab diff with a summary and line-level comments, with a free tier for open-source repos.
Herdr Dev Stack with Greptile
Add Greptile when parallel agents produce more diffs than one person can review: it indexes the whole repository so its comments carry context beyond the changed files, and it can be self-hosted for full control.
Herdr Dev Stack with Cursor Bugbot
Add Cursor Bugbot when a bug-focused reviewer is enough: it reviews GitHub or GitLab pull requests whatever wrote the code, though it bills through a Cursor account and plan rather than its own subscription.
CI/CD Add-ons
Add CI/CD when you want a dedicated pipeline for running tests, linting, or multi-stage builds before a deploy goes out. Many hosting platforms already redeploy automatically on every push on their own — a CI/CD tool adds the most value on top of that by gating the deploy on a passing test suite, and matters even more when the hosting choice does not auto-deploy at all, such as a self-hosted server.
Herdr Dev Stack with GitHub Actions
Add GitHub Actions when several agents are opening pull requests at once and every branch should be tested automatically before you spend time reviewing it.
Herdr Dev Stack with GitLab CI/CD
Add GitLab CI/CD for the same automatic testing when the repo lives on GitLab, with pipelines defined in a .gitlab-ci.yml file that can run on your own runners.
These are highlighted picks. To see all the tools, check the CI/CD Pipelines category.
Frequently Asked Questions about Herdr Dev Stack
Can agents keep working when my laptop is closed?
Only when Herdr runs on another machine. On a laptop, sessions survive closing the terminal window, but agents stop when the machine sleeps. On an always-on server, agents keep working with your laptop closed, and Herdr's remote mode attaches from your laptop over SSH with the full interface, panes and status included, while everything runs on the server. A phone reaches the same sessions through any SSH client, with a private network like Tailscale keeping the connection off the public internet. Herdr has no mobile app or hosted cloud. Size the server by agent count: 4 GB covers Claude Code alone, and several agents need more.
Do parallel agents in Herdr need separate worktrees?
If they work on the same repository, yes. Herdr manages terminal sessions, not code, so two agents editing the same checkout overwrite each other's changes. Each agent needs its own git worktree or clone, which some agent CLIs create for themselves and which otherwise sits outside Herdr. Orca and Conductor are the alternatives that manage worktrees for you.
Which agents can run which models?
Most combinations work, with varying effort. OpenCode and Pi take almost any provider's key, and Copilot CLI and Cursor CLI each offer a menu of Claude, GPT, and Gemini models through one subscription. Codex defaults to OpenAI but officially supports other providers through its configuration, local models included. Claude Code is built for Claude, and Anthropic supports only Claude models, though DeepSeek and GLM publish Anthropic-compatible endpoints some developers point it at.
How is Herdr different from tmux?
Both keep terminal sessions alive after you disconnect, and Herdr keeps tmux-style prefix keys, so the muscle memory carries over. The difference is agent awareness: Herdr recognizes the agent CLI running in each pane and shows whether it's working, blocked, idle, or done, and its socket API lets scripts and agents react to those state changes. It also supports the mouse for splitting and moving panes.
How is Herdr different from Orca?
Orca is a desktop app with a graphical diff review, built-in worktree management, and an embedded browser. Herdr is a terminal program: lighter, keyboard-driven, and at home on a remote server, but it leaves worktrees and code review to you. Developers who already work in tmux tend to prefer Herdr; those who want a GUI for reviewing parallel agents tend to prefer Orca.
Stacks Related to Herdr Dev Stack
Paseo Dev Stack
Developer
Run Claude Code, Codex, and other coding agents on your own machine with Paseo and steer them from any device.
Claude Code Dev Stack
DeveloperClaude Code in the terminal, desktop app, or web, paired with Claude and the add-ons to run it anywhere.
T3 Code Dev Stack
Developer
One open-source app for Claude Code, Codex, OpenCode, and other coding agents, on desktop and mobile.
Conductor Dev Stack
Developer
Run Claude Code, Codex, and Cursor's agent in parallel on your Mac with Conductor, each in its own workspace.
Scores
Popularity2/5
Herdr is a recently launched tool in a young category, known mainly among developers already running several terminal agents at once.
Learning Curve3/5
Prefix-key navigation will be familiar to tmux users, but running parallel agents safely means setting up worktrees by hand and learning each agent CLI, with models chosen as a separate step.
Flexibility5/5
Any of 20+ agent CLIs can run side by side, model-agnostic agents open the model layer to nearly any provider, and the socket API and plugin hooks let the workflow be scripted end to end.
Performance5/5
A single native Rust binary with no Electron runtime keeps overhead minimal even with many agent panes open, leaving the machine's memory for the agents themselves.
Portability5/5
Herdr is Apache 2.0-licensed, runs on macOS, Linux, and Windows, and works with any agent CLI and any git host, so every layer underneath can be swapped.
Tools in the Herdr Dev Stack Stack
Development Tools
Add-ons (optional — add any, or none)
Server
Remote Access
Terminal
Code Review
CI/CD
Herdr Dev Stack Pricing
Herdr itself is free and open source, so the bill comes from the agents. Claude Code starts at $20/month on Claude's Pro plan with no free tier, Codex has a limited free tier with Plus at $20/month, and both reach $200/month on their heaviest plans; OpenCode and Pi are free but bill per token with whichever provider you connect. Several agents usually means several subscriptions. A server is about €5.49/month for 4 GB and more for several agents; the terminal emulators and Tailscale's personal tier are free.
Apache 2.0-licensed single binary with no paid tiers.
Claude Code from $20/mo (no free tier); Codex free tier, Plus $20/mo; both up to $200/mo; Cursor $20/mo. One subscription per agent.
Model-agnostic agents bill per token with the provider you connect; open-weight models like DeepSeek and Kimi cost a fraction of premium ones.
Free tier covers individual devs; GitHub Team is $4/mo, GitLab Premium is $29/mo for more seats.
Ghostty, Alacritty, WezTerm, Kitty and iTerm2 are all free, open-source terminal emulators.
Tailscale's personal tier is free for up to 100 devices.
4 GB runs one agent, 8 GB or more suits several: Hetzner from €5.49, Hostinger $6.49 intro ($11.99 renewal), DigitalOcean $24–48; or a Raspberry Pi as a one-time cost.