---
id: dag-factory
version: "1.1.0"
license: Apache-2.0
license_treatment: permissive
maintenance: active
---

# dag-factory — Dynamically build Apache Airflow DAGs from YAML files

License: permissive · Maintenance: active · Downloads: 679.4K/mo

## What it is and what it does

dag-factory is a library that lets you define Apache Airflow DAGs using YAML configuration files instead of writing Python code. It reads YAML files and generates fully functional Airflow DAGs, making it possible to construct workflows without needing to understand Airflow's Python API or object model. The library supports scheduling, dynamically mapped tasks, custom operators, callbacks, and KubernetesPodOperator, covering most common workflow patterns.

It's designed for teams that want to avoid repetitive DAG boilerplate, enable non-Python developers to define workflows, or maintain DAG definitions in a more declarative, version-control-friendly format. The package depends on Apache Airflow, packaging, pathspec, pyyaml, and typer, and is actively maintained with support for modern Python versions (3.10–3.14) and Airflow 2.9+ including Airflow 3.

Use it for:

- Define and version-control multiple similar Airflow workflows using YAML templates without duplicating Python code.
- Enable data engineers or domain experts without Python expertise to create and modify Airflow DAGs.
- Rapidly prototype and iterate on workflow definitions by editing YAML files instead of redeploying Python modules.
- Manage large numbers of DAGs with consistent structure and defaults across an organization.
- Reduce maintenance burden by centralizing DAG logic in configuration rather than scattered Python files.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

dag-factory builds Apache Airflow DAGs declaratively from YAML configuration files, eliminating the need to write Python code for DAG construction.

Yes. dag-factory is actively maintained, has no known vulnerabilities, low install friction, and a permissive license. It solves a real problem—reducing boilerplate and enabling declarative DAG definition—for Airflow users. Install it if you manage multiple similar Airflow workflows or want to let non-Python developers define DAGs; skip it only if your DAGs are highly custom or require complex Python logic that YAML cannot express.

## Install

pip install dag-factory
uv add dag-factory
poetry add dag-factory

## Installing dag-factory

Before you install: Low install friction with a pure Python wheel. Actively maintained as of 2026-08-14 with 1452 repository stars. Requires Apache Airflow 2.9+ and Python 3.10+, both of which are current versions.

License in practice: Licensed under Apache License 2.0 (permissive), allowing commercial use, modification, and distribution with minimal restrictions—suitable for most production and proprietary projects.

Quickstart:

pip install dag-factory

from dag_factory.dag_factory import load_yml_dags

dags = load_yml_dags(dag_folder='dags', safe_mode=True)

Requires Apache Airflow 2.9+ and Python 3.10+ to be installed and configured; Airflow environment must be set up before dag-factory can load DAGs.

Verify before relying:

- Specific performance characteristics when handling large numbers of DAGs or complex task dependencies.
- Extent of custom operator support and any limitations on operator types that can be declared via YAML.
- Whether all Airflow 3 features are fully supported or if certain advanced features still require Python code.

## Package facts

- License: Apache-2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 679.4K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

airflow dag from yaml, declarative airflow dag builder, airflow dag factory configuration, yaml to airflow dag, airflow dag generation, airflow workflow as code yaml, airflow dag templating, airflow-plugin, workflow-orchestration, infrastructure-as-code

[View on SkillFed](https://skillfed.io/packages/dag-factory) · [View on PyPI](https://pypi.org/project/dag-factory/)