Herdr Crash Course: A Terminal Multiplexer for Running Many Coding Agents

Melvin Vivas · X video post · 2026-08-27 · 16:43 · 437 views · Open on X

Topics: AI Dev Tools & Productivity, AI Agents, Tool Use & MCP · Level: beginner

Summary

A beginner's walkthrough of Herdr, a terminal multiplexer built to run several coding agents at once and keep track of them. It covers installing Herdr, how workspaces, tabs and panes fit together, and how a multiplexer and its prefix key work. It also shows how to keep agents running on a VPS with herdr --remote, and how to let an agent control Herdr through its CLI using the official agent skill.

Key points

• Install Herdr with the copy-paste command on herdr.dev, then run herdr. Many workspaces, tabs and panes all run inside one terminal window.
• Suggested layout: one workspace per project and one tab per process or agent (for example implementer, reviewer, and npm run dev). Tabs can be split into vertical or horizontal panes.
• Herdr spots running agents (Pi, Tau and others) and lists them in a sidebar. Yellow means the agent is working and blue means it is done, with an optional notification. Notifications are off by default, so turn them on in settings.
• A multiplexer sits between your terminal and its subprocesses. They run in the background, so agents keep going after you close the app. Ctrl-B then Q detaches, and running herdr again re-attaches.
• The prefix key is Ctrl-B. Keys typed after the prefix control Herdr; keys without it go to the app inside (for example, Ctrl-D closes the agent). Shortcuts: Ctrl-B C opens a new tab, Ctrl-B ? lists shortcuts, Ctrl-B S opens settings.
• Remote work: running Herdr over SSH on the VPS adds lag, because the whole interface is streamed back to you. Instead run herdr --remote <ssh-host>. The interface stays on your machine and all processes run on the VPS, so you can turn your computer off.
• herdr session list shows local and remote sessions. herdr --remote <host> reconnects to a remote one.
• Herdr has a CLI that agents can use. Install the official Herdr agent skill (from the docs, with npx skills or by copying it into your skills folder) so an agent can create sessions, tabs and panes and start sub-agents.

Resources mentioned

• herdr · tool · x.com · check price
  A runtime that keeps real terminals open for coding agents on a local or rented machine, so agent sessions keep running when you disconnect.
  Also in: Terminal Tool Recommendation: herdr (Melvin Vivas on X · notes), herdr 0.9.0: Control Agents Across Multiple Machines (Melvin Vivas on X · notes), GPU devbox Docker image with coding agents on Runpod (Melvin Vivas on X · notes), GPU-Ready AI Devbox Docker Image on Runpod (PyTorch 2.8 + CUDA 12.8) (Melvin Vivas on X · notes) and 7 more
• Herdr documentation – Agent Skill · docs · herdr.dev · free
  Official Herdr docs page with the agent skill file that teaches agents to control Herdr through its CLI.
• tmux · tool · github.com · free
  Classic open-source terminal multiplexer that also uses a prefix key for shortcuts.
• cmux · tool · cmux.com · free
  Terminal tool for managing coding agents, which the speaker used before Herdr.
• Ghostty · tool · ghostty.org · free
  Terminal emulator used in the demo to host Herdr.
• Pi · tool · x.com · free
  A customizable coding agent that can be extended through its Extensions API and connected to several model providers.
  Also in: Sign in with ChatGPT: Setting Usage Limits for Each App (Melvin Vivas on X · notes), Pi reaches v1.0 (Melvin Vivas on X · notes), Claude Code mods: customize behavior and UI with plugins (Melvin Vivas on X · notes), Customizing Your Coding Setup with Pi Coding Agent Extensions (Melvin Vivas on X · notes) and 39 more
• Tau · tool · github.com · free
  Minimalist coding agent written in Python, a port of Pi's design, with native local-model support and extensions that can change its interface.
  Also in: Agent Harness Efficiency: Scaffolding Beats MCP vs. CLI (Melvin Vivas on X · notes)
• tau-herdr extension (by Ryan Dolphin) · tool · github.com · free
  Tau extension that adds Herdr integration.
• npx skills · tool · github.com · free
  Command-line tool for installing agent skills into your skills directory.
• llama.cpp · repo · github.com · free
  An open-source C/C++ engine for running GGUF models locally. Its llama-server command provides an OpenAI-compatible HTTP server.
  Also in: Running LLMs Locally Without an Expensive Rig (Melvin Vivas on X · notes), llama.cpp / Llama-macOS v0.5.0 release (Melvin Vivas on X · notes), Run llama.cpp GGUF Checkpoints in Hugging Face Transformers (Melvin Vivas on X · notes), llama.cpp v0.4.1 release announcement (Melvin Vivas on X · notes) and 28 more
• OpenCode · tool · x.com · free
  Open-source terminal coding agent that can be configured with different model providers, including open models.
  Also in: T3Code: a better interface for terminal coding agents (Melvin Vivas on X · notes), OpenCode Go: Open Coding Models (DeepSeek, Qwen, Kimi, GLM) for $10/mo (Melvin Vivas on X · notes), Trying OpenCode Go After Upgrading OpenCode (Melvin Vivas on X · notes), Orchestrator + Subagents in opencode with Open Models (Melvin Vivas on X · notes) and 10 more
• Hugging Face · website · x.com · free · recommended by both Bashiri Smith & Melvin Vivas
  Platform for hosting and finding ML models, datasets and papers. The quoted post says LocateAnything was trending there.
  Also in: Using an ML agent to train an open-source TTS model on your voice (Melvin Vivas on X · notes), Deploy Open-Source Models with Hugging Face Inference Endpoints (Melvin Vivas on X · notes), Deploying Qwen3.8 27B on Hugging Face Inference Endpoints (Melvin Vivas on X · notes), Using Hugging Face credits: Jobs, Inference Endpoints and Open Models (Melvin Vivas on X · notes) and 38 more

Try this

• Install Herdr from herdr.dev and start it with `herdr`.
• Set up one workspace per project and one tab per agent or process.
• Turn on notifications in settings (Ctrl-B S), since they are off by default.
• Use the mouse at first, check the key bindings with Ctrl-B ?, then move to keyboard shortcuts over time.
• Detach with Ctrl-B Q so agents keep running in the background.
• Run agents on a VPS with `herdr --remote <ssh-host>` instead of running Herdr over SSH, and check sessions with `herdr session list`.
• Install the official Herdr agent skill so your agents can control Herdr.
• Try Tau, especially if you like Pi, and install the tau-herdr extension.
• Build skills and workflows where one agent uses the Herdr CLI to start sub-agents in separate sessions or tabs and have them talk to each other.
• Set up a VPS where agents keep working after you turn your computer off, and check on them from your laptop or phone.