Skip to main content
Every request to Dataville names a source, the database you want the answer from. Sources range from Wikipedia and filings registries to research archives and package indexes, and all of them behave the same way. The same endpoints work for every source and every source returns the same JSON shape, so switching sources means changing only the source name and your keywords. The code around it stays the same. For the JSON itself, see Response format.

​
Browsing the catalog

Dataville’s catalog of sources is on the Data Library page. Each card shows what the source covers and its price per request, which depends on your plan. Click a card to open its search panel and run a test search before you write any code.
The Data Library page in the Dataville dashboard: category filters above a grid of source cards, beside a panel for running a test search.
The Data Library’s filters group sources as Financial, Academic & Research, Government Data, Reference, Food & Nutrition, and Developer. The Your Data filter holds the Google Sheets you’ve connected. If the source you need isn’t in the catalog, email support@dataville.com. We add new sources based on what developers ask for.

​
Searching a source

You can get a record out of a source with a keyword search, by pasting a link, or from an MCP tool in your editor or assistant. A search returns the most relevant result from the source you name, so search for a paper’s title, a company’s ticker, or a package name.

Search a data source

The endpoint, its parameters, and example responses.

Fetch an article by pasting its URL

Paste a link onto the API host and get the article back.

Connect an AI tool

The same search as an MCP tool in your editor or assistant.
To work with many records at once, query the source as a table.

​
Checking a license

Each source’s license is listed on the public source list. Most are public domain or openly licensed, and some require attribution when you display their content. Check the license before you show or redistribute results, and keep the source and metadata fields from each result so you can attribute it.

​
Downloading and querying sources

Most sources are also available as tables. Tables for sources Dataville fetches on demand hold only the records a lookup has already fetched, so they are smaller than the source itself. List available exports shows which sources are included.

Export datasets

Download a source as CSV or Parquet, or read its Parquet straight from DuckDB.

Query with SQL

Run read-only SQL across sources in Data Studio or over the API.