Skip to main content
Search for datasets on data.gouv.fr by keywords. This is typically the first step in exploring data.gouv.fr.

Parameters

query
string
required
Search keywords to find matching datasets.The API uses AND logic, so all words must match. Generic words like “données” or “fichier” are automatically filtered out to improve results. Use short, specific queries for best results.
page
integer
default:1
Page number for pagination.Results are paginated with 20 items per page by default. Use this parameter to navigate through multiple pages of results.
page_size
integer
default:20
Number of results per page.Maximum: 100. Controls how many datasets are returned in a single request.

Returns

Returns a formatted text response containing:
  • Total number of matching datasets
  • Current page number
  • For each dataset:
    • Title
    • Dataset ID
    • Short description (truncated to 200 characters)
    • Organization name
    • Tags (up to 5)
    • Number of resources (files)
    • Dataset URL
If no datasets match the query, returns a message indicating no results were found.

Usage notes

The search uses strict AND logic. All query words must appear in the dataset metadata. Generic stop words are automatically removed to prevent zero results.

Automatic query cleaning

The following words are automatically filtered from queries because they’re generic and often not present in dataset metadata:
  • données, donnee, donnees
  • fichier, fichiers
  • tableau, tableaux
  • csv, excel, xlsx, json, xml
If a cleaned query returns no results, the tool automatically retries with the original query.

Search tips

  • Use specific keywords related to the topic (e.g., “immobilier paris” instead of “données immobilier”)
  • Avoid generic words like “data” or “file”
  • Try singular and plural forms if you get no results
  • Use French keywords for better results on the French data platform

Example workflow

Typical workflow: search_datasetslist_dataset_resourcesquery_resource_data (or download_and_parse_resource for large files).
  1. Search for datasets: “prix immobilier”
  2. Get the dataset ID from results
  3. Use list_dataset_resources to see available files
  4. Use query_resource_data or download_and_parse_resource to access the data

Build docs developers (and LLMs) love