Data & API

The directory's metadata is open data in its own right. All of it is available as a download or over a read API, with no key and no account.

These exports describe the directory — what resources exist, who publishes them, where they are. They are not copies of the datasets themselves.

Downloads

/data/directory.json — The whole catalogue as one JSON document, with provenance.
/data/directory.csv — The same records, one row per resource, for spreadsheets.
/data/meta.json — Counts, provenance and the number of open validation warnings.
/data/search-index.json — The compact search index the site uses. Small shape, good for client-side search.

Current contents: 56 resources, 48 publishers, 23 categories, 237 tags, 164 endpoints. Every export carries the schema version, a generation timestamp and the repository commit it was built from, so a citation can be pinned to a specific revision.

Read API

Plain JSON over HTTPS, cached at the edge. No API key, no rate limit for ordinary use, and CORS is open so you can call it from a browser. Endpoints:

/api/v1/meta — Counts, schema version, generation timestamp and the commit the catalogue was built from.
/api/v1/resources — Search and filter resources. Accepts q, category, access, type, publisher, coverage, granularity, api, machine, sort, page, per_page.
/api/v1/resources/{id} — One resource, with its full endpoint list and metadata.
/api/v1/publishers — Every publisher with resource and category counts.
/api/v1/publishers/{id} — One publisher and the resources attributed to it.
/api/v1/categories — The controlled category vocabulary with live counts.
/api/v1/tags — Every tag in use, with counts.

For example, everything in the transport category that has an API:

curl 'https://data.stratum.my/api/v1/resources?category=transport&api=yes'

The API never proxies a publisher's dataset. It returns directory metadata and the original endpoint URLs, and you fetch the data from the publisher yourself, under their terms.

Licensing

The directory's own metadata — names, descriptions, categories, tags and links we wrote — is released under CC0 1.0. No permission needed, no attribution required (though it is welcome).

This does not relicense the underlying data. The datasets, APIs, reports and registries we point at belong to their publishers and are governed by the publisher's own terms. A CC0 export of our metadata tells you nothing about what you may do with the data you reach through it. Check each publisher.

Provenance and citing

Each export and API response carries a schema version, a generation timestamp and the source commit. To cite a specific state of the catalogue, pin to the commit:

{
  "schema_version": "1.0.0",
  "generated_at": "…",
  "commit": "…",
  "source": "https://github.com/Stratum-Technologies/Malaysian-Data-Directory"
}

Availability

The canonical catalogue lives in Git. Everything served here — the site, the API, the exports — is derived from it on deploy. If this service disappeared tomorrow, the directory would still exist in the repository, and anyone could rebuild it. That is deliberate.

Browse the source Improve the catalogue