dagu
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, runs on limited H/W resources. Alternative to Airflow / Cron / Job Scheduler.
Dagu is a self-hosted workflow orchestration engine with a built-in MCP server that exposes workflow inspection, run control, and Wiki management to authenticated AI clients. The MCP server runs at the `/mcp` endpoint alongside the Web UI and REST API, allowing agents to query workflow definitions, check run status, preview and apply DAG changes, and maintain documentation pages. It's designed for teams whose primary work isn't orchestration—keeping workflow structure in declarative YAML separate from business logic in existing scripts, containers, and commands. Dagu runs as a single binary with no external database, supporting Docker containers, SSH commands, Kubernetes Jobs, and human-task approvals.
Key Features
Use Cases
- 01AI-assisted workflow debugging and run inspection through MCP clients
- 02Agent-driven workflow documentation maintenance in Dagu's Wiki
- 03Programmatic workflow control and human-task completion from AI assistants
- 04Orchestrating ETL pipelines and data operations with AI-visible status and logs
- 05Automating infrastructure tasks (SSH commands, container jobs) with agent oversight
- 06Managing scheduled jobs and dependencies with AI-readable workflow history
Related MCP Servers
View moremarkitdown
Python tool for converting files and office documents to Markdown.
firecrawl
The context API to search, scrape, and interact with the web at scale. 🔥
prompts.chat
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
langflow
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
dagu — FAQ
What is the Dagu MCP server?+
The Dagu MCP server is a built-in endpoint (`/mcp`) that exposes workflow inspection, run control, and Wiki management tools to authenticated AI clients using the Model Context Protocol. It runs alongside the Dagu Web UI and REST API in the same process.
How do I enable the Dagu MCP server?+
The MCP server is automatically available at `http://localhost:8080/mcp` when you run `dagu start-all`. Connect MCP clients using the same authentication credentials configured for the Web UI (basic auth or JWT tokens).
Which MCP clients work with the Dagu MCP server?+
Any MCP-compatible client that supports HTTP transport can connect to the Dagu MCP server. This includes Claude Desktop, Cline, and other tools that implement the Model Context Protocol specification.
Do I need API keys or external services to use the Dagu MCP server?+
No external API keys or services are required. The MCP server uses Dagu's built-in authentication system (basic auth or JWT tokens). You only need to configure auth mode and credentials during initial Dagu setup.
Is the Dagu MCP server free to use?+
Yes, the built-in MCP server is part of the open-source Dagu engine licensed under GNU GPLv3. Community self-hosting requires no license key; licensed self-host features (SSO, RBAC, audit logging) require a separate license.
What can AI agents do through the Dagu MCP server?+
Agents can inspect workflow definitions and run history, check execution status, maintain Wiki documentation pages, preview and apply DAG changes, start/stop/retry runs, and complete human tasks—all through authenticated MCP tools.
How do I install dagu?+
Open the source repository on GitHub and follow its README. dagu is a mcp server — MCP Agents Market links you directly to the official repo.
Is dagu free?+
dagu is an open-source project hosted on GitHub. Check the repository for its license and any usage requirements.