herdrdev/herdr

the runtime your coding agents live on

View on GitHub ↗Jump to charts ↓Open shareable report

Summary Information

Updated 43 minutes ago
Added to GitGenius on July 2nd, 2026
Created on March 27th, 2026
Open Issues & Pull Requests: 241 (+1)
Number of forks: 2,371
Total Stargazers: 32,791 (+16)
Total Subscribers: 94 (-1)

Repository Insights (GitGenius)

Median issue/PR response: 2.9 hours
Mean response time: 3.5 days
90th percentile: 11.5 days
Tracked items: 322

How this project is maintained

Around half of the issues opened in the past year never receive a reply. 99% of open issues come from outside the core team, so the backlog reflects real-world use rather than internal planning. Almost all tracked open issues have seen activity in the last three months. Only 9% of issues opened in the past year have been closed. Three people close 79% of everything that gets resolved.

Charts & Analytics

Fetching additional details & charts...

Issue Activity (beta)

Open issues: 169
New in 7 days: 157
Closed in 7 days: 115
Avg open age: 9 days
Stale 30+ days: 0
Stale 90+ days: 0

Recent activity

Opened in 7 days: 131
Closed in 7 days: 95
Comments in 7 days: 189
Events in 7 days: 499

Top labels

  • maintainer-needed (151)
  • bug (129)
  • feature-req (91)
  • duplicate (67)
  • windows (32)
  • expected-behavior (25)
  • pending-release (22)
  • needs-decision (7)

Detailed Description

Herdr is a terminal-based agent multiplexer written in Rust that enables developers to run and manage multiple AI coding agents simultaneously within a single terminal interface. The project positions itself as a modern alternative to traditional terminal multiplexers like tmux, specifically designed with AI agent orchestration in mind. Rather than treating agents as generic processes, herdr provides native awareness of agent states, displaying whether each agent is blocked, working, done, or idle through visual indicators in a sidebar.

The core functionality centers on providing each agent with a real terminal environment rather than an emulated one, ensuring that full-screen terminal user interfaces render correctly. Users can organize agents across workspaces, tabs, and panes using mouse-native interactions or keyboard shortcuts with a ctrl+b prefix model. The interface supports clicking, dragging, and splitting panes, making workspace organization intuitive without requiring extensive keybinding memorization. A key differentiator from GUI-based agent managers is that herdr runs as a lightweight standalone binary within the terminal itself, with no Electron wrapper, no GUI application layer, and no telemetry or account requirements.

The project supports a comprehensive list of coding agents including Claude Code, OpenAI's Codex, Devin CLI, GitHub Copilot CLI, Cursor Agent, and many others. Detection works automatically through process-name matching and terminal output heuristics, with official integrations available for popular agents that enable native session restore and semantic state reporting. The supported agents table shows that most agents have full detection across idle/done, working, and blocked states, though some like Kiro CLI have partial support.

Herdr maintains session persistence through a background server that keeps agents running even when the user detaches from the terminal. This enables reattachment from any terminal, including over SSH from mobile devices. The remote functionality allows running herdr on a VPS and connecting from a local terminal while preserving features like image pasting that plain SSH plus tmux would break. The binary is approximately 10MB and runs on Linux and macOS with Windows in beta, requiring no external dependencies.

The project includes a local Unix socket API that allows agents themselves to orchestrate herdr, creating workspaces, splitting panes, spawning helpers, and subscribing to state changes. This bidirectional integration enables agents to drive the multiplexer programmatically rather than just being passive occupants of panes. Custom integrations can be written in any language using the socket API and CLI.

The repository overlaps contributors with major projects including Electron, ClickHouse, and PyTorch, suggesting involvement from experienced systems developers.