Topics / #task-orchestration

Task-orchestration projects on GitHub

As of , the most-starred Task-orchestration projects on GitHub are kestra (29.7K stars), celery (29K stars) and luigi (18.8K stars). GitHub.Locker tracks 6 Task-orchestration repositories with 78K stars combined. The fastest-growing on October 16 was kestra, with +29 stars in a day.

6Projects
78KStars in total
+42Stars in a day

Hot right now

Most stars gained in a day · Oct 16

The all-time top

6 projects · by stars

  1. #1 · kestra-io/kestra

    kestra

    Kestra is an open-source, event-driven orchestration platform for building data, AI, and infrastructure workflows. You describe them in declarative YAML and manage everything from a built-in web UI.

    29.7K
  2. #2 · celery/celery

    celery

    Celery is a distributed task queue for Python: clients put units of work (tasks) on a queue and dedicated worker processes monitor the queue and run them — across threads, machines, even datacenters.

    29K
  3. #3 · spotify/luigi

    luigi

    Luigi is a Python package for building complex pipelines of batch jobs. It handles dependency resolution, workflow management, visualization, failure handling and command line integration.

    18.8K
  1. 4

    ordewell

    202ordewell/ordewell · TypeScript

    Multi-agent task orchestration for coding agents: you describe one goal and it becomes an ordered plan of tasks, each executed by its own Claude Code, Codex or OpenCode session, then verified.

  2. 5

    agent-control

    195loulanyue/agent-control · Python

    Agent Control is a production-grade control plane and orchestration engine for autonomous AI agents, handling distributed task dispatching, heartbeat leasing, DAG workflow execution and data-crawling pipelines, with a built-in MCP server for LLM clients like Claude Desktop and Cursor.

  3. 6

    oh-my-subagents

    191ringlochid/oh-my-subagents · Python

    A local orchestration runtime for Codex and Claude subagents: durable AI teams with persistent task state, reusable responsibility trees and interruption recovery, instead of ad-hoc delegation.

Often tagged together