LargitData Developer Portal

LargitData runs two kinds of programmable surface. The Public Content API is free, read-only and needs no credentials: it returns the same product catalog, knowledge-base index and blog that this site publishes as HTML. The commercial data APIs — the Social Media API and the Cross-border Ecommerce Data API — are contracted services authenticated with an API key. Everything you need for the first is on this page; the second starts with a trial key request.

Quickstart

No sign-up, no key, no SDK. One request:

curl -s https://www.largitdata.com/api/v1/products | jq '.data[].id'

Start from the service index to discover everything that is available:

curl -s https://www.largitdata.com/api/v1/

The machine-readable description of every endpoint below lives at /openapi.json (YAML). Point your generator at it:

npx @openapitools/openapi-generator-cli generate \
  -i https://www.largitdata.com/openapi.json \
  -g python -o ./largitdata-client

Endpoints

Base URL: https://www.largitdata.com/api/v1/ — all endpoints are GET only (HEAD and OPTIONS are also answered) and readable from any origin via CORS.

lang accepts zh-tw (default), en or ja. page_size ranges from 1 to 100 and defaults to 50.

Products in the catalog

Response format

Every successful response uses the same envelope. Collections carry meta and links; single items carry data only.

{
  "data": [ { "id": "infominer", "name": "InfoMiner ...", "url": "https://..." } ],
  "meta": { "total": 9, "page": 1, "page_size": 50, "total_pages": 1, "lang": "en" },
  "links": {
    "self": "https://www.largitdata.com/api/v1/products?page=1",
    "next": null,
    "prev": null,
    "documentation": "https://www.largitdata.com/developers/"
  }
}

Paginate by following links.next until it is null. Responses are cached for 10 minutes and carry an X-API-Version header.

Errors and error codes

Errors are never HTML. Every failure under /api/ returns RFC 9457 application/problem+json with a stable code and a hint describing how to recover:

{
  "type": "https://www.largitdata.com/developers/#error-invalid_parameter",
  "title": "Bad Request",
  "status": 400,
  "code": "invalid_parameter",
  "detail": "page_size=9999 is outside the supported range.",
  "instance": "/api/v1/pages",
  "hint": "page_size must be an integer between 1 and 100.",
  "documentation_url": "https://www.largitdata.com/developers/",
  "parameter": "page_size"
}

Requests to a non-API path that does not exist return a real HTTP 404. Send Accept: text/markdown and you get a short markdown document listing where to look next instead of a styled HTML page.

Machine-readable resources

Everything an agent needs to navigate this site without scraping it:

Commercial APIs and authentication

Two LargitData services deliver data under contract rather than from this site. They are authenticated with an X-API-Key request header:

curl -H "X-API-Key: $LARGITDATA_API_KEY" https://<your-assigned-host>/...

To request a trial key, email [email protected] with the platforms, keyword or category scope, expected request volume and target delivery format. We reply with a trial key, the endpoint documentation for your scope, and integration support.

Fair use and support

• The Public Content API is not rate-limited today. Please stay under roughly 1 request per second and set a descriptive User-Agent so we can contact you before changing anything that affects you.
• Responses are cacheable for 10 minutes; honour Cache-Control rather than polling.
• Content returned by the API stays subject to the site terms. Attribution to LargitData is required when you republish it.
• Breaking changes get a new version prefix (/api/v2/); /api/v1/ keeps its contract.