deploying-airflow

Deploys Airflow DAGs and projects. Use when deploying Airflow or answering anything about deployment - deploying DAGs/projects, pushing code, setting up CI/CD, deploying to production or deployment strategies for Airflow.

1.36x

Quality

—

Does it follow best practices?

Impact

97%

1.36x

Average score across 3 eval scenarios

Securityby

Risky

Do not use without reviewing

Evaluation results

93%

22%

Automating Deployments for a Growing Airflow Project

Astro CI/CD pipeline with branch-optimized deploy strategy

Criteria

Baseline

With context

astro deploy --dags flag used

100%

astro deploy --image flag used

100%

Full astro deploy on main

100%

DAG-only deploy on feature branches

100%

Main deploy condition correct

50%

100%

Separate image and DAG jobs

100%

83%

No full deploy on feature branches

25%

100%

Astro UI CI/CD mentioned

62%

Secret used for API token

100%

Deployment ID referenced

100%

75%

100%

15%

Deploy Airflow 3 on Kubernetes for the Data Platform Team

Airflow 3 Kubernetes Helm deployment with git-sync

Criteria

Baseline

With context

gitSync enabled

100%

gitSync repo field

100%

gitSync branch field

100%

gitSync subPath field

100%

gitSync wait field

100%

apiServer key used

100%

dagProcessor enabled

100%

helm repo add command

100%

helm repo update

100%

helm install namespace flags

100%

helm upgrade command

100%

41%

Local Airflow Development Environment

Airflow 3 Docker Compose local dev setup

Criteria

Baseline

With context

API server service

100%

Standalone DAG processor

100%

Correct image tag

60%

100%

LocalExecutor configured

100%

AIRFLOW__ variable pattern

100%

Connection via AIRFLOW_CONN_

80%

100%

DAGs volume mount

100%

Prerequisites documented

100%

Service descriptions

42%

100%

Repository: astronomer/agents
Commit: 8827e93

Evaluated: 8 days ago
Agent: Claude Code
Model: Claude Sonnet 4.6

Table of Contents

Automating Deployments for a Growing Airflow Project Local Airflow Development Environment Deploy Airflow 3 on Kubernetes for the Data Platform Team

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.