Skip to main content
fastCRW Pergamon

Every source,
one API key.

Official APIs, public registries and statistics offices, reached the same way as a scrape: one key, one credit wallet, one error shape. Every response carries the licence and the attribution the source asks for, so you can use what you get.

Named after the library of Pergamon, whose scribes gave us parchment and kept collecting.

464
Providers
Official APIs and registries
4,396
Capabilities
Endpoints you can call today
28
Categories
From packages to statistics
1
API key
One wallet, one error shape
[ 01 / 03 ]  ·  catalog
//
The catalog
//

What you can call today.

Every provider here is live. Switch one off and it leaves this page and the API together.

Showing 9 of 464 providers

Government · AF

openAFRICA

open.africa

Africa's largest central repository for Government, Civil Society, Corporate and Donor Agency Data.

datasetdataset-namesdatasetsdataset-suggest+11
Public records · AU

AusTender OCDS

api.tenders.gov.au

Every Australian Government contract notice published on AusTender, in Open Contracting Data Standard form - buyer and supplier names with ABNs, contract value, period, UNSPSC category, procurement method and the full amendment history.

amended-betweencontract-noticeending-betweenpublished-between+1
Research · EU

AlphaFold Protein Structure Database

alphafold.ebi.ac.uk

EMBL-EBI and Google DeepMind's AlphaFold Protein Structure Database: predicted three-dimensional structures for over 200 million proteins, searchable by UniProt accession, AlphaFold model id, amino-acid sequence or sequence MD5, returning per-model confidence (pLDDT) and interface scores, downloadable mmCIF/PDB/BCIF/PAE files, predicted protein complexes, and AlphaMissense pathogenicity predictions for every possible missense variant of a human protein.

alphamissensechecksum-summarycomplexesprediction+2
Developer

crates.io

crates.io

Rust crate metadata from the official registry: every published version, downloads, dependencies and owners.

categoriescategorycratedependencies+8
Health · CA

Canadian Nutrient File

food-nutrition.canada.ca

Health Canada's Canadian Nutrient File, the national reference food composition database: nutrient values per 100 g for about 5,700 foods commonly eaten in Canada, with household serving-size conversion factors, refuse and cooking yield factors, and the full nutrient dictionary behind them, in English and French.

foodfoodsnutrient-amountsnutrient-group+11
Finance · CA

Bank of Canada Valet

www.bankofcanada.ca

Canadian exchange rates, policy rates and every other Valet series.

group-detailgroup-observationsgroupssearch-groups+4
Statistics · AU

Australian Bureau of Statistics

data.api.abs.gov.au

Australian official statistics over SDMX: CPI, labour force, population and every other published dataflow.

categoryschemecategoryschemescodelistcodelists+6
Development

Agify.io

api.agify.io

Predicts a person's age from their first name using a name-frequency model, returning an estimated age, the sample count behind it and the name.

predict
Geo

ArcGIS public GIS layers

www.arcgis.com

Esri's public ArcGIS Online catalogue and the hosted feature services behind it - millions of openly shared GIS layers, maps, scenes and government open data published by agencies, universities and companies worldwide - searchable by keyword, tag, owner and bounding box, and queryable down to individual features and aggregate statistics.

feature-layerfeature-queryfeature-query-geometryfeature-service+21
455 more to go
[ 02 / 03 ]  ·  how it works
//
How it works
//

Three parts, no SDK.

01

Discover

GET /v1/data lists every live provider, its capabilities, the parameters each one takes and what it costs. It is free to call.

GET /v1/data
02

Call

One GET per capability. Required parameters come from the catalog, so there is nothing provider-specific to learn.

GET /v1/data/npm/package
  ?name=react
03

Attribute

Every response carries a provenance block: the source, the exact upstream URL, the licence and the attribution line to display.

"attribution_text":
  "Source: npm registry"
A whole response
{
  "success": true,
  "provider": "ecb",
  "capability": "series",
  "credits_used": 1,
  "request_id": "fP4Nxx4Urn-H06AxscbM_",
  "provenance": {
    "source": "European Central Bank",
    "upstream_url": "https://data-api.ecb.europa.eu/service/data/EXR/D.USD.EUR.SP00.A",
    "licence": "ESCB policy on the reuse of ECB statistics ...",
    "attribution_text": "Source: ECB",
    "retrieved_at": "2026-10-11T16:44:02.117Z",
    "cache_age_seconds": 0,
    "is_stale": false
  },
  "data": { ... }
}
[ 03 / 03 ]  ·  licensing
//
Licensing
//

We read the terms first.

A source only appears here once someone has read what it permits. That answer ships with the data.

The licence travels with the response

Not a page in our docs: the actual licence text and attribution line for that provider, on every call, so you can satisfy it programmatically.

Nothing ships un-reviewed

A provider whose terms nobody has read is not reachable. There is no default-on path onto this page.

Caching follows the licence

Where a source says do not retain, we do not cache. Where it is public domain, we cache hard and you get the latency back.

Want a source that is not here? Tell us which one. The catalog grows by request.