Dagu

Dagu is a self-hosted workflow engine for teams that want to define jobs in YAML and operate them through a web interface. It is a strong fit for teams combining existing scripts and services; its main trade-off is a free plan capped at two API keys, with administrative controls reserved for paid licenses.

Overview

Dagu runs as a single binary, locally or across distributed workers, without an external database or message broker. That keeps deployment requirements modest, while the built-in interface brings workflow runs, logs, and retries into one place. The GNU GPL v3.0 license makes the core software open source, and the product sits within Workload Automation Software.

Key features

Workflows are defined as DAGs in YAML and can call shell scripts, Docker containers, SSH commands, SQL queries, HTTP requests, chat completion steps, and Agent Harness steps. This breadth is useful when a team wants to orchestrate existing tools rather than rewrite each task for a single execution model.

Cron scheduling supports timezones, catchup windows, and overlap policies. Retry policies, zombie detection, lifecycle handlers, and notifications add ways to handle failures and run conditions; dependencies and event triggers extend workflows beyond fixed schedules. Cross-platform execution is supported.

Built-in actions include HTTP, mail, SSH, SFTP, Git, S3, Redis, LLM, browser, computer, SQL, PostgreSQL, SQLite, and DuckDB. Dagu also supports MCP connections from clients including Claude Code, Codex, Cursor, VS Code, Gemini CLI, OpenCode, Zed, and Cline. This gives teams multiple integration paths, though the range of actions will matter most to users whose workflows need those systems.

Paid licenses add SSO, role-based access control, audit logging, incident routing, and support to self-hosted servers. Those are meaningful operational controls for teams managing access and incident response; smaller deployments may be able to stay on the free plan if two API keys are sufficient.

Pricing

The Community plan is free forever and includes unlimited servers and workers, but allows up to two API keys. That is generous for self-hosted execution capacity, though the key cap may constrain teams integrating multiple clients or services.

Team costs $50 / mo or $500 / yr, saving two months on annual billing. It includes three self-host Dagu server licenses, unlimited workers, SSO (OIDC), RBAC, audit logging, standard support, and incident routing. It suits teams needing administrative controls across a limited number of servers.

Pro costs $150 / mo or $1,500 / yr, also saving two months annually. It raises the allowance to 15 self-host server licenses while retaining unlimited workers and the Team controls. It is the clearer fit for organizations running Dagu across more servers; the higher price buys server capacity rather than additional listed controls.

Enterprise has custom pricing, a custom licensed server allotment, and priority support. A free self-host license trial lasts 14 days and requires no credit card. Standard support comes with Team and Pro, while community help through documentation, GitHub Issues, and Discord is best effort; support is handled directly by Dagu's maintainer. Hosted service data is on Google Cloud Platform, with encryption in transit and at rest, access controls, and tenant isolation safeguards.

Platforms

Dagu is available across API, Linux, macOS, self-hosted, web, and Windows environments. Its single-binary deployment and distributed-worker model make it relevant to both local use and multi-worker setups.

Who it's for

Dagu is best suited to teams that want self-hosted workflow orchestration, YAML-defined DAGs, and a web interface for run management without operating a separate database or message broker. It is also a plausible choice for teams that need to connect scripts, containers, databases, and service actions in one workflow. Organizations requiring more than two API keys, SSO, RBAC, audit logging, or incident routing should budget for a paid license.

Pros and cons

• Pros: A single-binary deployment without an external database or broker lowers infrastructure requirements.
• Pros: The free plan offers unlimited servers and workers, useful when execution capacity matters more than API-key count.
• Pros: Broad execution actions and scheduling controls accommodate varied workflow steps and failure handling.
• Cons: The Community plan's two-key limit may be restrictive for integrations spanning several clients or services.
• Cons: SSO, RBAC, audit logging, and incident routing require Team or Pro, so the free license lacks the listed administrative controls.

Alternatives

Consider Kestra if a free, Apache 2.0-licensed option better suits your licensing preference. Rundeck is another freemium choice, with a free Community plan intended for small teams and requiring registration.

Apache Airflow is worth considering if you want a no-cost option released under the Apache License. Windmill may suit teams comparing a freemium workflow platform with a free tier that includes workspace, user, and storage allowances.

Luigi is a free Apache-2.0 licensed Python package. Prefect may fit readers who prefer a freemium service with a Hobby tier capped at two users, one workspace, five deployments, and 500 minutes per month of Prefect Serverless.

HCL Workload Automation is a paid alternative with a free trial. StackStorm is another free, open-source option with no paid products offered by its project.

Verdict

Choose Dagu if you want self-hosted YAML workflows with broad execution actions and a generous free allowance for servers and workers. Its lean deployment model is a persuasive reason to start with it; look elsewhere or move to a paid tier if the two-key cap or the need for administrative controls makes Community insufficient.