# PrefectHQ/prefect

Prefect is a workflow orchestration framework for building resilient data pipelines in Python.

Repository: https://github.com/PrefectHQ/prefect
Canonical: https://ross.abutalabs.com/products/prefect
Homepage: https://prefect.io
Language: Python
License: Apache-2.0
License Family: permissive
Topics: python, workflow, data-engineering, data-science, workflow-engine, prefect, infrastructure, ml-ops, data-ops, automation, orchestration, data, observability, pipeline
Last push: 2026-08-26T23:21:59+00:00

## Health v2 (maintenance only)
Score: 95/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 100
- inputs: {"age_days": 2987, "days_push": 7, "days_rel": 8, "gap_med": 5, "n_releases_24m": 144}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 23693, forks 2483 (observed 2026-08-28T04:11:35.208590+00:00)

## What it is
Prefect is an open-source workflow orchestration framework that turns Python functions into production-grade data pipelines using decorators like @flow and @task. It provides scheduling, retries, caching, state tracking, event-driven automations, and a real-time UI, with a self-hosted server or managed Prefect Cloud for monitoring deployments.

## Use cases
- orchestrate data pipelines in python
- replace cron jobs with scheduled workflows
- add retries and caching to python scripts
- schedule and monitor etl jobs
- build ml training and inference pipelines
- trigger workflows from external events
- self-host a workflow orchestration engine
- run data workflows on kubernetes

## When to choose
- you want to elevate python scripts into production workflows with minimal code
- you need dynamic, data-driven pipelines without yaml or dsls
- you want built-in scheduling, retries, caching, and observability out of the box
- you need to run workflows anywhere python runs, from local to kubernetes
- you prefer a pythonic alternative to airflow

## When to avoid
- you need a language-agnostic orchestrator for non-python codebases
- your workflows are simple enough for plain cron or shell scripts
- you require strict dag definitions defined statically before runtime
- you want a fully managed service without any self-hosted component

## Facets
- artifact type: framework
- maturity: stable
- function: workflow-automation, scheduling, monitoring, etl, developer-tools
- domain: data-science, machine-learning, developer-tools
- platform: python, cross-platform, self-hosted, cloud
- tags: orchestration, data-pipelines, mlops, data-ops, observability, durable-execution, event-driven, retry-logic, task-caching, airflow-alternative, data-engineering, automation, docker, kubernetes

## Member repositories
- PrefectHQ/prefect (main) score 95

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:35.208590+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 2026-08-29T16:56:22.050419+00:00, confidence not recorded.
  - readme: https://github.com/PrefectHQ/prefect (fetched 2026-08-28T04:11:35.208590+00:00, sha 5b50f2873412)
  - homepage: https://prefect.io (fetched 2026-08-29T07:54:10.809775+00:00, sha 933c451bb69e)
  - site_page: https://docs.prefect.io/v3/get-started/index (fetched 2026-08-29T07:54:10.814538+00:00, sha 5eea312bf7f4)
  - site_page: https://www.prefect.io/pricing (fetched 2026-08-29T07:54:10.812598+00:00, sha 5d09c51bc062)
  - site_page: https://www.prefect.io/changelog (fetched 2026-08-29T07:54:10.816395+00:00, sha 864c0ccb22be)
  - site_page: https://www.prefect.io/company (fetched 2026-08-29T07:54:10.818232+00:00, sha 166be5a8ae7b)
- Data as of 2026-08-30T08:39:29.467469+00:00.