#345 · Primary category: AI Agents & Automation

dagu

agentic-workflows ai-agents ai-workflows airflow-alternative cron dag data-pipeline devops durable-execution human-in-the-loop job-scheduler llm mcp mcp-gateway mcp-server self-hosted task-automation workflow-engine workflow-orchestration workflow-scheduler

Self-hostable workflow orchestrator for teams whose main work isn't orchestration. Declarative YAML over your scripts, SSH commands, containers, etc; keep workflows separate from business logic. One binary, no database. Alternative to Airflow / Cron / Job Scheduler.

Project last updated:08/29/26

GitHub Stars

3.8K

Forks

318

Contributors

89

License

GPL-3.0

Why we included this project

Dagu is for teams whose day-to-day work is scripts, runbooks, and on-prem jobs, and who would rather not stand up a data platform just to get scheduled execution. It wraps plain shell commands, Docker containers, Kubernetes jobs, and remote SSH commands into DAG-style workflows defined in declarative YAML, then handles scheduling, retries, human approvals, and run history for you. The whole thing is a single binary with a web UI and no external database or message broker, so it can run in an air-gapped or edge environment right where your credentials and data already live. What makes it interesting for AI-assisted operations is native MCP support and the ability to launch external coding-agent CLIs inside a workflow when a step genuinely needs a model. Teams that want Airflow-style orchestration without the heavy footprint, plus the occasional agent in the loop, get one tool that spans both.

Articles for this project

No articles for this project yet.

To suggest a topic or contribute an article, contact us.

Related projects in this category