Steampipe renders SQL as the main query language for more than purely databases, Cloud infrastructure included. Let's find out how.
The notion behind Steampipe is that SQL has been the data access standard for decades. It levels the playing field, easily integrates with other systems and accelerates delivery. So why not leverage it for things other than the database, like querying APIs and Cloud services?
Steampipe does just that; it enables a uniform data access method for querying:
Cloud providers like AWS, Azure, GCP, Cloudflare, Alibaba Cloud, IBM Cloud, and Oracle Cloud
Cloud-based services likeGitHub, Zoom, Okta, Slack, Salesforce, and ServiceNow
Structured files like CSV, YML and Terraform
Ad hoc investigation of network services like DNS & HTTP
Kubernetes configurations, manifest files, and helm charts
Use SQL to execute commands on local or remote hosts
The question is how does it pull this feat off? Under the hood Steampipe leverages PostgreSQL and the Foreign Data Wrappers to present the various data sources as database tables. The Steampipe FDW does not directly interface with external systems, but instead relies on plugins to implement the API/provider specific code and return it in a standard format via gRPC. Thus all the Postgres-specific logic is encapsulated in the FDW, and API and service specific code resides only in the plugin which translate the APIs to SQL-queryable tables.
As such there's a myriad of plugins for any case imaginable, like:
OpenAPI Use SQL to instantly query resources from OpenAPI.
steampipe plugin install openapi
Query all the endpoints available for an API:
select api_path, operation_id, summary, deprecated, tags from openapi_path;
AWS Use SQL to query infrastructure including servers, networks, identity and more from AWS.
select title, create_date, mfa_enabled from aws_iam_user
Hacker News Steampipe marshals the HN API data into queryable tables letting you interactively explore it via the command line interface or your favorite SQL client.
steampipe plugin install hackernews
select score, descendants as comments, title from hackernews_top where type = 'story' and lower(title) like '%sql%' order by score desc;
CSV Query data from the users.csv file:
select first_name, last_name from users;
There's the plugin repository where you'll find plugins for the following categories:
AI
Asset Management
IaC
Internet
Media
Network
PaaS
Public Cloud
SaaS
Security
Software Development
Virtualization
Website Monitoring
There's 146 plugins in total at the time of writing but the list keeps growing.
Then install a plugin for your favorite service (e.g. AWS, Azure, GCP, GitHub, Kubernetes, Hacker News, etc).
What's more, Steampipe latest architecture decouples Steampipe from its batteries-included Postgres database, so you could use plugins with your own database. That effort resulted in three new plugin flavors: for Postgres, for SQLite, and for standalone export.
A great tool with a great ecosystem. SQL for the win!
This month's TIOBE Index is out and the headline asks the question "Where is SQL Going?" Its chart provides the answer of "down", a fate that also applied to Assembly Language. The language that is mo [ ... ]
Continuing the series highlighting conferences you may have missed we come right up to date with Spring I/O and KotlinConf, both of which took place during May 2025.
