Naver Search results API (NAVER_SERP) - querying.ai

NAVER_SERP returns one page of results from Naver’s web documents tab, 15 documents per page. With "action": "page" it reads one naver.com page (a blog post, cafe article, news story, Knowledge iN answer or any other naver.com page) and returns its text.


`taskType`	`NAVER_SERP`
Input	`query` (required for search); `page`, `country`; or `action: "page"` with `url`, `maxChars`
Credits	1 per task
Country	Defaults to `KR`.
Location	Not supported (`422 LOCATION_UNSUPPORTED`).

Search

curl -X POST https://api.querying.ai/v1/async/task \
  -H "Authorization: Bearer $QUERYING_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "taskType": "NAVER_SERP", "payload": { "query": "running shoes for flat feet", "page": 2 } }'

Field	Required	Description
`query`	Yes	The search query, 1-100,000 characters. `prompt` is accepted when `query` is absent.
`page`	No	Results page, an integer from 1 to 10. Defaults to `1`. Other values return `422 VALIDATION_ERROR`.
`country`	No	Defaults to `KR`, Naver’s market.

For example, page 2:

{
  "organicResults": [
    { "position": 16, "title": "...", "link": "https://...", "displayedLink": "example.com›path", "snippet": "...", "page": 2 }
  ]
}

Field	Description
`organicResults`	The documents on the page. `position` counts across pages, so the first document on page 2 is `16`. Each row carries its `page`.
`notice`	Present when Naver withholds the results behind its age check: `{ "type": "age_verification_required", "scope": "organicResults", "message" }`. `message` quotes Naver when it can be read.

An empty organicResults in a completed task means Naver itself returned no documents.

Read a naver.com page

curl -X POST https://api.querying.ai/v1/async/task \
  -H "Authorization: Bearer $QUERYING_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "taskType": "NAVER_SERP", "payload": { "action": "page", "url": "https://blog.naver.com/example/223000000000", "maxChars": 10000 } }'

Field	Required	Description
`action`	Yes	`page`.
`url`	Yes	An `https` address on `naver.com` or one of its subdomains. Other addresses return `422 VALIDATION_ERROR`.
`maxChars`	No	Caps `text`, an integer from 1,000 to 50,000. Defaults to 20,000.

For example:

{
  "type": "blog",
  "url": "https://blog.naver.com/example/223000000000",
  "title": "...",
  "text": "...",
  "author": "...",
  "publishedAt": "...",
  "images": ["https://..."],
  "truncated": false
}

Field	Description
`type`	`blog`, `cafe`, `news`, `terms`, `kin` or `generic`.
`title`, `text`	The page’s title and body text.
`author`, `publishedAt`, `cafeName`	Present when the page has them.
`images`	Up to 30 image addresses from the body.
`truncated`	`true` when `text` was cut at `maxChars`.

These fail the task with PAGE_UNAVAILABLE and a reason: a redirect that leaves naver.com, a missing page (404 or 410), a cafe article that only members can read, and a page with no readable body.

Limits and errors

NAVER_SERP ignores markdown, rawResponse and searchQueries. It rejects html and analysis with 422 and citationAnalysis with 400; see unsupported flags.
Other request errors are listed in Errors.

Naver AI Brief, which also returns the organic documents under its summary
Naver AI Tab
Regions and location