run-kit

run-kit is an open-source tmux orchestrator and remote management console developed by Sahil Ahuja as part of the shll.ai toolkit. It operates as an agent-agnostic harness, exposing active tmux sessions and panes through a browser-based dashboard accessible from a local workstation or mobile devices over networks like Tailscale. In daily use, run-kit provides terminal multiplexing workflows specifically suited for parallel development. The `run-kit riff` command provisions isolated agent workspaces across distinct Git worktrees (integrating with sibling tool `wt`), while `run-kit serve` launches a live web interface to view session outputs, host metrics, and pane states. Because it connects directly to tmux without wrapping application processes, run-kit is completely model- and agent-agnostic—equally supporting Claude Code, Aider, custom shell scripts, REPLs, and local development servers. What distinguishes run-kit from agent harnesses that wrap LLM APIs is its pure terminal multiplexer foundation. Rather than binding to a single agent protocol or proprietary execution loop, run-kit treats all AI coding agents as standard shell processes. This design ensures that agent orchestration remains durable even as underlying AI tools and CLI agents evolve. run-kit is free, open source under the MIT license, and distributed as a component of the broader shll terminal ecosystem.

Tags: agent harness, ai agent, cli, developer tools, terminal

Visit run-kit

Key features

  • Agent-agnostic tmux orchestration — Exposes live tmux panes and sessions directly to a browser console without wrapping or modifying internal agent protocols. - Remote and mobile web dashboard — Launches a web UI with host metrics and a multi-session sidebar via run-kit serve, optimized for remote inspection over Tailscale. - Worktree workspace generation — Spawns parallel coding agent sessions into isolated Git worktrees using run-kit riff (integrated with wt). - shll ecosystem compatibility — Integrates seamlessly with companion tools such as tu for session cost tracking and wt for Git branch isolation. - Status and notification hooks — Provides status indicators and alerts to track agent progress and pane state transitions.

Use cases

Use Case: Parallel Agent Execution Across Worktrees A developer dispatches several coding agents to work concurrently on separate features or bug fixes. Using run-kit riff, each agent runs inside an independent Git worktree, allowing the developer to supervise multiple tasks simultaneously from the run-kit sidebar.

Use Case: Remote Session Monitoring via Mobile A developer initiates a long-running agent task on an office workstation and checks live terminal outputs, agent prompts, and host performance metrics from a phone or tablet over Tailscale.

Use Case: Unified Multitasking Console A developer manages backend microservices, database REPLs, and AI CLI agents within a single tmux session, relying on run-kit's web console as a lightweight window into all running processes.

Who it is for

run-kit is designed for command-line developers who rely on tmux and want remote or parallel monitoring for CLI-based AI coding agents across Git worktrees. It is not suitable for developers seeking a standalone desktop IDE or an agent harness that requires a proprietary LLM API client wrapper.

Install guide

Installation Install run-kit alongside the rest of the shll toolkit via the official installer: bash curl -fsSL https://shll.ai/install | sh

First Run Verify that tmux and git are installed on your path, then start the web dashboard: bash run-kit serve Navigate to the local URL displayed in the terminal output (or via your Tailscale address) to access the console.

Example Session To spawn an agent or command in an isolated Git worktree, execute riff: bash run-kit riff feature-refactor Switch to the dashboard running under run-kit serve to monitor the new worktree pane alongside other active tmux sessions.

Part of CLI Coding Agents

Categories: AI Coding Agents, Developer Tools, Productivity

Related CLI agents

  • Repomon — Repomon is an open-source developer harness and repository companion tool created by Ali Hamza Azam. Built with a Rust core…
  • Even — Even is an agent-native desktop workspace and orchestration environment developed by Todience, Inc. It integrates a native terminal, an embedded…
  • SWE-agent — SWE-agent is an open-source autonomous software engineering agent harness built and maintained by researchers from Princeton University and Stanford University.…
  • Claude Squad — Claude Squad is an open-source terminal user interface (TUI) orchestrator and multi-instance session harness for Claude coding agents, developed by…
  • Amazon Q Developer CLI — Amazon Q Developer CLI is an open-source command-line coding assistant developed by Amazon Web Services (AWS). Built primarily in Rust,…
  • Warp — Warp is an AI-integrated terminal emulator and CLI workspace built by warpdotdev. Engineered in Rust with native GPU rendering, Warp…
  • AI Tools
  • Categories
  • Industries
  • CLI Coding Agents
  • MCP Servers
  • MCP Categories