# apache/airflow

Apache Airflow - A platform to programmatically author, schedule, and monitor workflows

Repository: https://github.com/apache/airflow
Canonical: https://ross.abutalabs.com/products/airflow
Homepage: https://airflow.apache.org/
Language: Python
License: Apache-2.0
License Family: permissive
Topics: airflow, apache, apache-airflow, python, scheduler, workflow, automation, dag, data-engineering, data-integration, data-orchestrator, data-pipelines, data-science, elt, etl, machine-learning, mlops, orchestration, workflow-engine, workflow-orchestration
Last push: 2026-08-27T00:12:37+00:00

## Health v2 (maintenance only)
Score: 98/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 97, longevity 100
- inputs: {"age_days": 4160, "days_push": 7, "days_rel": 21, "gap_med": 14.5, "n_releases_24m": 37}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 46613, forks 17684 (observed 2026-08-28T04:12:11.095536+00:00)

## What it is
Apache Airflow is an open-source platform for programmatically authoring, scheduling, and monitoring workflows as directed acyclic graphs (DAGs) defined in Python. It provides a web UI, scheduler, and a rich ecosystem of operators and hooks for integrating with cloud services and data tools.

## Use cases
- schedule daily ETL pipelines that extract and load data into a warehouse
- orchestrate machine learning training and deployment workflows
- replace cron jobs with a monitored, dependency-aware scheduler
- build data integration pipelines across AWS, GCP, and Azure services
- trigger and monitor backfills of historical data processing
- coordinate multi-step batch jobs with retries and alerting

## When to choose
- you need to orchestrate complex, dependency-driven batch workflows in Python
- you want a mature, battle-tested scheduler with a web UI and large community
- your pipelines integrate with many external systems via existing operators and hooks
- you need dynamic pipeline generation and Jinja templating

## When to avoid
- you need low-latency, event-driven streaming rather than scheduled batch workflows
- your use case is simple cron-style scheduling without dependencies or monitoring needs
- you want fully managed orchestration without operating scheduler infrastructure yourself

## Facets
- artifact type: framework
- maturity: stable
- function: workflow-automation, scheduling, etl, streaming, monitoring, developer-tools
- domain: data-science, machine-learning, big-data
- platform: python, self-hosted, cli, cloud
- tags: workflow-orchestration, dag, data-pipelines, scheduler, elt, mlops, data-integration, airflow, data-engineering, automation, devops, docker, kubernetes, web-server

## Member repositories
- apache/airflow (main) score 98

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:11.095536+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:21:36.545518+00:00, confidence not recorded.
  - readme: https://github.com/apache/airflow (fetched 2026-08-28T04:12:11.095536+00:00, sha 4cd70bbc2c77)
  - homepage: https://airflow.apache.org/ (fetched 2026-08-29T07:45:03.776002+00:00, sha 0fd640dfe92c)
  - site_page: https://airflow.apache.org/docs (fetched 2026-08-29T07:45:03.785272+00:00, sha 18d076f22859)
  - site_page: https://airflow.apache.org/docs/stable/start.html (fetched 2026-08-29T07:45:03.787408+00:00, sha cdec3054d4fc)
- Data as of 2026-08-30T08:39:29.467469+00:00.