herdr is a terminal runtime for coding agents that keeps agent terminals alive in a background server and combines local and remote machines in a single window. Each pane is marked as working, blocked, or idle so you can track multiple agents at a glance.
Project overview
It addresses a concrete pain point for agent-driven development: coding agents no longer die when you close a terminal or lose SSH, and a single Rust binary with no Electron dependency runs in the terminal you already use while existing agents like Claude Code, Codex, Cursor, OpenCode, and Grok drive it via a CLI and socket API.
Project type
AI Agent · AI Coding
Use cases
Coding & Development · Automation
Deployment
Refer to project documentation
License
Apache-2.0
Best for
Developers and AI engineers who run coding agents such as Claude Code, Codex, Cursor, OpenCode, or Grok and need sessions that survive disconnection
Anyone managing multiple agent terminals across local and remote machines who wants a single-window view with per-pane status
Key capabilities
Agents drive herdr through the CLI and socket API to spawn panes, prompt each other, and wait until another agent is blocked.
Getting started
Install via the curl script or brew install herdr, then run herdr where the work lives.
Detach with ctrl+b q and reattach later with herdr.
Evidence and sources
GitHub project description: the runtime your coding agents live on
README: herdr keeps terminals running in a background server when you close the client or lose your SSH connection.
README: never hunt for the stuck one — every pane is marked working, blocked, or idle.
README: agents drive herdr through the cli and socket api: they can spawn panes, prompt each other, and wait until another agent is genuinely blocked.
README: runs what you already run — claude code, codex, cursor, opencode, grok and the rest. herdr doesn't wrap or replace them; it owns their terminals.