Orchestrator.inc

Orchestrator.inc released agent-orchestrator v0.14.10, an Apache-2.0 licensed, self-hosted tool that coordinates multiple AI coding agents from a single Kanban board, with 14,175 GitHub stars and a last commit one hour before publication. The tool uses a two-level structure in which a main orchestrator agent splits a goal into pieces, starts a worker for each piece in its own isolated Git worktree, and routes failed CI checks and review comments back to the agent that owns the branch, supporting 38 agents including Claude Code, OpenAI Codex, Cursor, Aider, Goose, Gemini CLI and GitHub Copilot. Orchestrator runs locally and free with no account, requires users to bring their own API keys so billing stays with each provider, and offers a free iPhone, iPad and Android companion app that pairs by QR code.

Open Source Alternative to: - Stars 14,175 - Last commit 1 hour ago - Repository age 9 months - Version v0.14.10 - License Apache-2.0 https://openalternative.co/licenses/apache-2 - Self-hosted Yes https://openalternative.co/collections/self-hosted - Repository agent-orchestrator https://github.com/OrchestratorInc/agent-orchestrator Orchestrator sits above your AI coding agents https://openalternative.co/categories/developer-tools/ai-assisted-coding/ai-coding-agents and manages the chaos that comes with running several at once. Instead of juggling terminal windows and GitHub tabs, you get one board that shows every agent, its branch, its pull request, and what it needs from you. The core idea is a two-level structure. A main orchestrator agent breaks a goal into pieces, starts a worker for each piece in its own isolated Git worktree, and routes feedback back to the right agent automatically. Workers never step on each other's branches. - Kanban board, placed by facts. Lanes reflect real state: what the agent is doing, CI results, review status. Nothing needs manual updating. - CI and review routing. When a check fails or a reviewer leaves a comment, Orchestrator sends it back to the agent that owns that branch, with the log or thread attached. - Agent reviewer. A built-in reviewer agent reads every PR before a human does and returns findings to the same worker session. - 38 supported agents. Claude Code, OpenAI Codex, Cursor, Aider, Goose, Gemini CLI, GitHub Copilot, and more. Each project picks its own worker, orchestrator, and reviewer agents independently. - Chat and native terminal. Work with an agent through structured chat or open its own TUI in the same worktree. Switch between them without losing context. - Mobile companion. The free iPhone, iPad, and Android app pairs by QR code. Get notified when an agent asks a question, reply in chat, or approve a plan without opening your laptop. It runs locally, free, with no account required. You bring your own API keys; Orchestrator doesn't proxy calls, so billing stays with each provider. Cloud plans add hosted sandboxes so agents keep running after the laptop closes. Compared with other AI coding agent orchestrators https://openalternative.co/categories/developer-tools/ai-assisted-coding/ai-coding-agent-orchestrators , Orchestrator's focus is the feedback loop: failed CI goes to the right agent, review threads return to the session that wrote the code, and the board surfaces only what actually needs a human decision. The desktop app runs on macOS, Windows, and Linux. Every Sunday we deconstruct one proprietary app and pick the best open source alternatives worth switching to. - Stars - Last commit - License AI Run multiple AI coding agents in parallel, side by side - Stars Stars growing quickly over the last 30 days117,694 - Last commit 53 minutes ago - License MIT Agent Development Environment that runs Claude Code, Codex, Gemini, and other CLI agents in parallel across isolated git worktrees, with terminals, browser, diff review, and git tools built in. Compare with AI-powered coding made simple and efficient - Stars 49,470 Desktop and phone app for your Claude, Codex and Grok sessions - Stars 24,522 Run coding agents from your phone, desktop, or terminal - Stars 23,305 Run dozens of AI coding agents in parallel on your machine - Stars 16,022 Dashboard for running multiple coding agents in parallel - Stars 6,207 An agentic IDE that can control and automate itself - Stars 4,948 Code with Claude, Codex, Gemini, and more in one workspace - Stars 2,185 Parallel AI agent workspaces with Git and project context - Stars 1,397 Run Claude Code, Codex, Grok, and more from one app Stars growing quickly over the last 30 days Last commit 51 minutes ago License MIT Advanced AI coding assistant that enhances development workflow with intelligent code suggestions, automated debugging, and seamless integration. Orchestrates Claude Code, Codex, and Grok through a single desktop and mobile app, reusing the AI subscriptions you already pay for. Self-hosted orchestrator for coding agents like Claude Code, Codex, and Cursor, controlled from phone, desktop, web, or CLI. Stars growing over the last 30 days Desktop app for macOS and Linux that orchestrates multiple AI coding agents simultaneously, each in its own isolated Git worktree, so you can work on many tasks at once. Desktop app that runs 25+ coding agents simultaneously, each isolated in its own Git worktree, so you can orchestrate code changes without switching tools. Local-first IDE that runs multiple AI coding agents in parallel, supports plugins built in a single prompt, and lets any script or bot spawn work threads. Desktop coding workspace that connects to AI subscriptions you already have, with parallel agents, worktrees, one-click PRs, and multi-project management. Run multiple AI coding agents in isolated Git worktrees, with cross-repo context, GitHub/Linear integrations, and support for Claude, Codex, Cursor, and more. - Stars Stars growing quickly over the last 30 days126 - Last commit 3 days ago - License Apache-2.0 A Mac app that brings multiple AI coding agents into a single window, running on your existing subscriptions with parallel agent sessions, PR management, and a built-in browser.