Wikidata
The structured knowledge base behind Wikipedia: entity search and key facts for any person, place, organisation or thing.
- Provider id:
wikidata - Tools: 2, at $0.001 per call
- Categories: Knowledge & Reference
- Homepage: www.wikidata.org
- Provider docs: www.wikidata.org/wiki/Wikidata:Data_access
- Licence: CC0 1.0
- Attribution: Wikidata (wikidata.org)
| Tool | Price | What it does |
|---|---|---|
wikidata/entity | $0.001 | Key facts about one Wikidata entity: what it is, country, dates, place, people, population, website. |
wikidata/search | $0.001 | Find Wikidata entities (Q-ids) by name, with label and one-line description to tell them apart. |
Inputs are JSON bodies, and unknown fields are rejected with 422 invalid_input (not charged). POST /v1/inspect with a tool's id returns the same input schema, plus the output schema.
Wikidata Entity Facts
wikidata/entity · $0.001 per call (local) · Knowledge & Reference
Key facts about one Wikidata entity: what it is, country, dates, place, people, population, website.
Reads one item (Q-id) and returns its label, description, aliases, Wikipedia link and a curated set of about 30 common properties (instance of, country, inception, birth/death, headquarters, CEO, population, area, official website, image…), with item values resolved to labels and their Q-ids kept for follow-ups. Only the best-ranked values are kept (up to 5 per property). It does not run SPARQL or list every property; for prose use wikipedia/summary. Find the Q-id with wikidata/search.
Related: wikidata/search, wikipedia/summary.
Input
| Field | Type | Required | Description |
|---|---|---|---|
id | string | yes | A Wikidata item id, e.g. 'Q7251'. Pattern ^[Qq][1-9][0-9]{0,11}$. |
lang | string | no | Label language, e.g. 'en', 'de', 'ja'. Pattern ^[a-z]{2,3}(-[a-z]{2,8})?$. Default "en". |
Example
{
"id": "Q7251"
}node akashi.mjs run wikidata/entity --input '{"id":"Q7251"}'Wikidata Entity Search
wikidata/search · $0.001 per call (local) · Knowledge & Reference
Find Wikidata entities (Q-ids) by name, with label and one-line description to tell them apart.
Matches labels and aliases (prefix match, so 'Alan Tur' works) and returns each candidate's Q-id, label and description, e.g. to tell Paris the city from Paris Hilton. Pass the chosen id to wikidata/entity for its facts. It matches names, not facts: 'tallest building in Japan' will not work; use akashi/answer or wikipedia/search for questions.
Related: wikidata/entity, wikipedia/search.
Input
| Field | Type | Required | Description |
|---|---|---|---|
query | string | yes | A name or label, e.g. 'Alan Turing' or 'Kyoto'. 1–250 characters. |
limit | integer | no | From 1 to 20. Default 5. |
lang | string | no | Label language, e.g. 'en', 'de', 'ja'. Pattern ^[a-z]{2,3}(-[a-z]{2,8})?$. Default "en". |
Example
{
"query": "Alan Turing",
"limit": 3
}node akashi.mjs run wikidata/search --input '{"query":"Alan Turing","limit":3}'