andyrewlee/awesome-agent-orchestrators

What it solves

A single reference point for the rapidly growing ecosystem of open-source tools that coordinate multiple AI agents. Instead of hunting through GitHub, this list groups tools by their role (parallel coding, autonomous loops, swarms, personal assistants, infrastructure) so you can quickly compare options for running, supervising, and chaining agents in your own workflows.

How it works

The project is a curated GitHub list (an "Awesome" list). Each entry links to a specific tool's repository and includes a one-sentence description that highlights its key feature or supported agent CLIs. The tools are organized into seven categories, each addressing a distinct orchestration pattern: terminal-based parallel coding agents (TUI/CLI), desktop and web dashboards, multi-agent swarms that split tasks, autonomous loop runners that persist toward a goal, task runners that pull from issue trackers, reusable infrastructure and primitives, and always-on personal assistants. A companion website (agent-orchestrators.com) offers a richer web view.

Who it’s for

Developers and teams who want to run multiple AI coding agents concurrently, build autonomous agent pipelines, or embed an orchestration layer into their own tooling. It’s also useful for anyone exploring the agent ecosystem to understand what categories exist and how the tools differ.

Highlights

• Covers 7 distinct orchestration patterns, from terminal multiplexers to mobile-ready dashboards.
• Each entry includes the agent CLIs it supports (e.g., Claude Code, Codex, OpenCode) so you can filter by your stack.
• Includes a dedicated section for infrastructure primitives for building your own orchestrator.
• Links to a companion web version with more detail per tool.