Announcing the Astronomer Registry

  • Pete DeJoy

Today, we’re excited to announce The Astronomer Registry: a discovery hub for Airflow providers designed to bridge the gap between the Airflow community and the broader data ecosystem.


The Airflow providers framework has long been a valuable asset for the community. DAG authors benefit from access to a deep library of standard patterns while maintaining code-driven customizability; it enables a tight integration between Airflow and almost any tool in the modern data stack.

But it doesn’t stop there. The value of the provider framework also extends beyond the Airflow user to the service provider itself.

The data ecosystem loves Airflow, as it allows service providers to fit their tools into highly-contextual pipelines that help enable broader business outcomes. An easy plug-and-play architecture allows folks to use tools the way they’re meant to be used, which produces a more favorable outcome for all downstream data users.

So we’ve made it easy for the data ecosystem get involved with the Airflow community. With a sample repository and support from the Astronomer team, anyone, from a major technology company to an individual developer writing DAGs, can publish a provider on the Registry from any public repository. The Registry aggregates and curates the best bits of the ecosystem so that developers can easily discover and access the integrations they need.


Providers and modules

The providers framework is a robust solution for interfacing with a large catalog of tools, but we found that a user’s first exposure to the framework can be overwhelming. Across hooks, operators, sensors, transfers, and providers, there are many different entities designed to do different things. To simplify the experience, we’ve introduced the concept of Modules to make the logical separation clear:


Easy navigation

The Registry’s goal is to consolidate only the most crucial information a user needs to get started with a provider package and its modules, including access instructions, required parameters, and example DAGs. Our goal is to help users easily conceptualize how providers and modules are used contextually in DAGs.

Note that Registry documentation does not conflict or compete with core Airflow documentation; in fact, the docs rendered on the Registry are built from the same source as the Airflow docs! Instead, the Registry amalgamates Airflow docs, example DAGs, and quick access instructions from both the core Airflow repository and third-party repositories to improve discoverability and access across the entire ecosystem.

Through many user interviews, it became clear that discoverability is a major pain point for the Airflow community. To address that pain, we built the Astronomer Registry with first-class, intuitive search across both providers and modules, making it easy to find and implement anything in the ecosystem without spending hours reading source code.

Partner Providers [Beta]

Many companies have reached out asking for assistance in creating an Airflow Provider for their tool. The Registry allows partners in the data ecosystem to deliver high-quality providers and modules to the Airflow community more quickly, with success metrics baked in.

We’re proud to launch a beta program with a small batch of partners who have worked closely with us to refine the build and release process, including Great Expectations and Fivetran. We look forward to delivering many more of these packages in the coming months and working to bring more first-class integrations to the Airflow community.

Please visit our sample provider repo for more info and feel free to get in touch if you’d like to publish a provider package.

Looking Ahead

There are a few items that we’re particularly excited about working towards with the Registry in coming months:

We sincerely hope you enjoy what we’ve built.

If you’re interested in becoming a beta tester for any of the above features, please don’t hesitate to reach out to us at

Ready to Get Started?

Get Started Free

Try Astro free for 14 days and power your next big data project.