{"name":"RimZ: Realtime Dashboard for Agentic Coding with tmux and Zellij","description":"RimZ is a powerful realtime dashboard and control room designed for agentic coding, enabling humans and AI agents to collaborate seamlessly within tmux or Zellij environments. Built with Rust, it provides comprehensive observability, orchestration, and automation capabilities for managing fleets of coding agents. This tool enhances productivity by offering real-time insights, message-based steering, and scriptable workflows for AI-driven development.","github":"https://github.com/rimio-ai/rimz","url":"https://osrepos.com/repo/rimio-ai-rimz","source":"osrepos.com","sourceDescription":"This repository profile is provided by osrepos.com, an open source repository discovery platform.","repositoryProfile":"https://osrepos.com/repo/rimio-ai-rimz","generatedFor":"open source discovery and AI-assisted research","markdown":"https://osrepos.com/repo/rimio-ai-rimz.md","json":"https://osrepos.com/repo/rimio-ai-rimz.json","topics":["agent-orchestration","agentic-ai","ai-agents","coding-agent","rust","terminal-ui","tmux","zellij"],"keywords":["agent-orchestration","agentic-ai","ai-agents","coding-agent","rust","terminal-ui","tmux","zellij"],"stars":null,"summary":"RimZ is a powerful realtime dashboard and control room designed for agentic coding, enabling humans and AI agents to collaborate seamlessly within tmux or Zellij environments. Built with Rust, it provides comprehensive observability, orchestration, and automation capabilities for managing fleets of coding agents. This tool enhances productivity by offering real-time insights, message-based steering, and scriptable workflows for AI-driven development.","content":"## Introduction\n\nRimZ is a cutting-edge realtime dashboard and control room for harnessing agentic coding, built on tmux and Zellij. It's designed to facilitate seamless collaboration between a human developer and a fleet of AI agents within a single terminal room. RimZ provides an immediate overview of every agent, displaying its live status, task, context health, and real-time cost, with a sidebar that intelligently routes your attention to where it's most needed.\n\nThis lightweight Rust-based binary integrates directly into your existing tmux or Zellij setup, preserving your keybinds and allowing official agent CLIs to run untouched. RimZ lays the foundation for advanced harness engineering and loop engineering, offering primitives like a robust sidebar for observability, a unified command grammar for various agents, durable messages for steering, supervised runs for scripting, and event-driven wakeups for automation.\n\n## Why Use and Key Benefits\n\nRimZ offers a unique set of features that streamline agentic coding workflows, making it an indispensable tool for developers working with AI agents:\n\n*   **Extremely Lightweight Integration:** Runs as a single binary within your familiar terminal multiplexer, ensuring zero learning curve and compatibility with existing tools and official agent applications.\n*   **Realtime Harness Dashboard:** Gain immediate insights into your agent fleet, including working state, task, model usage, effort, context health, live token statistics, and dollar costs.\n*   **Intelligent Attention Routing:** The sidebar triages your fleet, highlighting agents that require your input or attention, allowing you to quickly navigate to the relevant pane.\n*   **Worktrees for Every Agent:** Launch agents in isolated Git worktrees with dynamic layouts, enabling parallel development on different features without conflicts.\n*   **Cross-Model Agent Teams:** Combine the strengths of various AI models and providers into cohesive teams, assigning roles for planning, coding, or reviewing to optimize outcomes.\n*   **Message-Driven Agent Communication:** Interact with agents using a Slack-like messaging system, ensuring durable message delivery, respecting context windows, and facilitating agent-to-agent communication.\n*   **End-to-End Scriptability:** Integrate agents into scripts, CI pipelines, and automated workflows with `rimz agents -p`, offering supervised turns, exit codes, and JSON output for robust automation.\n*   **Engineer Your Own Loops:** Schedule agent runs based on time, intervals, cron expressions, or as watchdogs that trigger on command results, enabling hands-off automation.\n*   **Auto-Continue While Away:** Agents automatically recover from rate limits and transient API errors, resuming work without manual intervention, allowing you to step away with confidence.\n*   **Remote Control from Your Phone:** Answer agent prompts and steer your fleet directly from official Claude and ChatGPT mobile apps, maintaining workflow continuity from anywhere.\n*   **Continuous Local or Remote Work:** Connect to your RimZ room from any location via SSH, with self-healing links that ensure your session persists and agents continue working seamlessly.\n\nRimZ is currently beta software, actively developed and dogfooded by its creators, ensuring it's ready for daily personal use while continuously evolving.\n\n## Installation\n\nInstalling RimZ is straightforward, with options for various environments.\n\n**Using the Install Script:**\n\nThe recommended way to install RimZ is via the official install script:\n\nsh\ncurl -fsSL https://raw.githubusercontent.com/rimio-ai/rimz/main/scripts/install.sh | sh\n\n\nOn macOS, this script automatically uses Homebrew if available.\n\n**Homebrew (macOS/Linux):**\n\nIf you prefer using Homebrew:\n\nsh\nbrew install rimio-ai/rimz/rimz\n\n\n**Cargo (Rust Package Manager):**\n\nFor Rust developers, you can install RimZ directly from crates.io:\n\nsh\ncargo install --locked rimz\n\n\nAfter installation, the first `rimz` run will offer to install necessary hooks for agents to report to the room. You can preview these changes with `rimz hooks install --dry-run`.\n\n## Examples\n\nHere are some common ways to get started and leverage RimZ in your daily workflow:\n\n**Opening a Room and Launching Agents:**\n\nNavigate to your project directory and open a RimZ room, then launch your preferred agents:\n\nsh\ncd ~/code/query-engine\nrimz\nclaude\ncodex\n\n\n**Working with Agent Teams and Worktrees:**\n\nLaunch multiple agents in isolated worktrees with dynamic layouts:\n\nsh\nrimz agents claude,codex --worktree=feat-x       # Claude + Codex, side by side\nrimz agents 'vim,claude+term' --worktree=feat-y  # editor, agent, shell in one tab\nrimz teams forge -w feat-complex                 # Use a predefined team for complex features\n\n\n**Steering Agents with Messages:**\n\nCommunicate with your agents, addressing them by their handle:\n\nsh\nrimz message @claude \"add coverage for the expiry edge cases\"\nrimz message @coder --wait \"did the migration land? one line\"\ngit diff main | rimz message @reviewer --stdin \"review this\"\n\n\n**Automating Tasks with Scripting and Loops:**\n\nIntegrate agents into scripts or set up automated loops:\n\nsh\nrimz agents codex \"Prepare the release checklist.\" -p --timeout 30m --output-format json\nrimz loop add watchdog --check \"cargo test\" --on fail \\\n    --agent codex --prompt \"fix the failing test\" --every 15m\n\n\n**Remote Access:**\n\nConnect to your RimZ room from anywhere:\n\nsh\nrimz remote add dev dev-box:~/code/query-engine\nrimz remote connect dev          # Rebuilds the room, agents where you left them\nrimz remote connect dev --web    # Access the same room in your browser\n\n\n## Links\n\nFor more in-depth information and to contribute to the project, refer to the official documentation and repository:\n\n*   **GitHub Repository:** [https://github.com/rimio-ai/rimz](https://github.com/rimio-ai/rimz){:target=\"_blank\"}\n*   **Documentation Index:** [https://github.com/rimio-ai/rimz/blob/main/docs/README.md](https://github.com/rimio-ai/rimz/blob/main/docs/README.md){:target=\"_blank\"}\n*   **Installation Guide:** [https://github.com/rimio-ai/rimz/blob/main/docs/guide/installation.md](https://github.com/rimio-ai/rimz/blob/main/docs/guide/installation.md){:target=\"_blank\"}\n*   **Design Principles:** [https://github.com/rimio-ai/rimz/blob/main/DESIGN.md](https://github.com/rimio-ai/rimz/blob/main/DESIGN.md){:target=\"_blank\"}\n*   **Architecture Overview:** [https://github.com/rimio-ai/rimz/blob/main/ARCHITECTURE.md](https://github.com/rimio-ai/rimz/blob/main/ARCHITECTURE.md){:target=\"_blank\"}\n*   **License (MIT):** [https://github.com/rimio-ai/rimz/blob/main/LICENSE](https://github.com/rimio-ai/rimz/blob/main/LICENSE){:target=\"_blank\"}","metrics":{"detailViews":2,"githubClicks":1},"dates":{"published":null,"modified":"2026-09-17T23:31:14.000Z"}}