---
id: prefect-aws
version: "0.7.11"
license: Apache License 2.0
license_treatment: permissive
maintenance: active
---

# prefect-aws — Prefect integrations for interacting with Amazon Web Services.

License: permissive · Maintenance: active · Downloads: 6.0M/mo

## What it is and what it does

prefect-aws is an integration layer that bridges prefect workflow orchestration with AWS services. It provides reusable blocks (configuration objects), tasks, and worker infrastructure to simplify common patterns—file I/O with S3, credential management via Secrets Manager, serverless execution with Lambda, container orchestration on ECS, and batch/ETL jobs via Batch and Glue. Rather than calling boto3 directly, you use prefect-aws's pre-built components, which handle authentication centrally, add observability through logging and state tracking, and include built-in retry logic for transient failures.

The package depends on prefect as its orchestration engine and boto3/aiobotocore for AWS API access. It supports Python 3.10–3.13 and is actively maintained. Most users install via `pip install "prefect[aws]"`. Setup involves creating an AwsCredentials block with access keys, then using service-specific blocks in workflows.

Use it for:

- Upload and download files from S3 within workflows without managing boto3 sessions directly.
- Deploy long-running workers on ECS for production data pipeline execution at scale.
- Invoke Lambda functions synchronously or asynchronously as part of a workflow.
- Retrieve and manage secrets stored in Secrets Manager with centralized credential handling.
- Run ETL jobs on Glue or submit batch compute jobs via Batch from a flow.
- Authenticate to RDS PostgreSQL databases using IAM tokens for passwordless access.

## Worth the install?

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

Provides pre-built blocks, tasks, and workers for integrating Prefect workflows with AWS services including S3, Lambda, Secrets Manager, ECS, Batch, and Glue.

Yes. The package is actively maintained, has low install friction, carries a permissive license, and integrates cleanly with prefect for AWS workflows. Choose it if you are already using prefect and want to reduce boilerplate around AWS service calls; skip it if you are not using prefect or prefer direct boto3 usage.

## Install

pip install prefect-aws
uv add prefect-aws
poetry add prefect-aws

## Installing prefect-aws

Before you install: Low install friction with a pure-Python wheel distribution. Actively maintained with a release 9 days old and recent commits. Depends on prefect, boto3, aiobotocore, and several utility libraries—all widely used and stable.

License in practice: Licensed under Apache License 2.0 (permissive). No restrictions on commercial use, modification, or distribution; suitable for proprietary projects.

Quickstart:

pip install "prefect[aws]"

from prefect_aws import AwsCredentials
from prefect_aws.s3 import S3Bucket

creds = AwsCredentials(
    aws_access_key_id="KEY",
    aws_secret_access_key="SECRET",
    region_name="us-east-2"
)
bucket = S3Bucket(bucket_name="my-bucket", credentials=creds)

Requires an AWS account with appropriate IAM permissions for the services you intend to use.

Verify before relying:

- Whether RDS IAM authentication plugin is production-ready or remains experimental as stated.
- Specific performance characteristics or throughput limits for batch operations.
- Whether cross-region operations are fully tested and recommended for production use.

## Package facts

- License: Apache License 2.0 (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 6.0M/month (top 5,000 on PyPI)
- Known vulnerabilities: none known

## Tags

prefect aws integration, aws workflow orchestration, s3 lambda blocks, ecs worker deployment, secrets manager integration, batch job processing, glue etl workflows, aws-integration, workflow-orchestration, infrastructure-as-code

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