Basics
What is a data source?
In PromptQL, a data source is anything you can ask questions about.
That includes traditional databases like Postgres and MySQL, but also REST APIs, GraphQL endpoints, and even custom functions or services. If it returns data, it can be a data source.
How do I connect my data (and what happens next)?
We use native data connectors to make your data accessible through PromptQL. These connectors are designed for specific systems: relational databases, NoSQL, REST APIs, GraphQL APIs, and more.
Once your data sources are connected, PromptQL acts as the interface. You ask a question in plain language, and PromptQL figures out what data it needs, pulls the right pieces from one or more sources, and assembles an accurate, reliable result.
You don't need to worry about where the data lives — PromptQL can traverse across your connected sources to answer your question.
You can get started by choosing one of the supported connectors:
- Amazon Athena
- Amazon Redshift
- BigQuery
- Databricks
- MySQL
- PostgreSQL
- Snowflake
- TypeScript Lambda
- Python Lambda
- Go Lambda
Building your own connector
If you're working with a custom data source, you can create your own connector. The NDC Spec describes how this works, and we offer SDKs and a full tutorial to guide you.
- Tutorial
- SDKs: