Naver Scraper API – 네이버 Blog, News, Place & Map Reviews, KiN

Naver API and scraper in one Actor (네이버 블로그·뉴스·플레이스·지식iN 데이터): public Naver data, no login, no cookies, no API key. Naver blog search and full blog posts, Naver news search, most-viewed ranking, full articles and news comments, Naver Place (Naver Map) search, place details and visitor reviews, KiN (지식iN) Q&A, Naver web search results, autocomplete suggestions, DataLab Shopping Insight keyword ranks, Naver Webtoon lists and Korean stock quotes.

• Array inputs: many keywords, URLs or place ids in one run; each input is isolated (one bad URL never fails the run).
• Agent-ready output: flat JSON, the same keys on every row (null when unknown), lean / table modes for fewer tokens, free error rows with a fix, a summary row last.
• English in, Korean out: common English terms are translated (Gangnam restaurants → 강남 맛집); each row shows queryUsed.
• Monitoring: deltaKey returns only new reviews, comments, posts or articles since the last run.
• Pay per row from $0.0005 (reviews, comments). Free dry run, free capabilities manifest, maxCostUsd stop.

Leave the input empty to get today's Naver News most-viewed ranking (about 1 s).

What you get

You want	Operation	One row =
Naver blog posts for a keyword	blog_search (+ include: ["fullText"])	title, blog, date, snippet, URL (+ full text, images)
Naver news + comments	news_search, news_ranking, news_articles, news_comments	article, press, time, body; comment text, likes, time
Naver Place / Naver Map reviews	place_search, place_details, place_reviews	place, rating, review counts, hours, menu, booking link; review text, rating, visit date
KiN Q&A (지식iN)	kin_search, kin_questions	question, answers, views
Naver search results (SERP)	web_search, autocomplete	rank, title, URL, domain, snippet; suggestion
Keyword demand	shopping_keywords	DataLab Shopping Insight rank, keyword, category
Webtoons, stocks	webtoons, stock_quotes	title, author, rating; price, change, volume, PER

Prices (pay per event)

Event	Price	Rows
search-result	$0.002	Naver blog, news, KiN, web search rows; blogger RSS posts
post	$0.003	Full blog post, full news article, KiN question with answers
place	$0.001	Naver Place row (search result or details)
review	$0.0005	Naver Place visitor review or news comment
keyword-row	$0.002	Autocomplete suggestion, Shopping Insight keyword
trend	$0.001	News ranking, webtoon, stock quote row
Actor start	$0.00005	per run

Error, summary, plan (dry run), capabilities and insights rows are free.

Research workflow in one call (keyword → articles → full text → comments → insights): {"operation":"news_search","keywords":["삼성전자"],"include":["fullText","comments","insights"],"maxItemsPerInput":10,"commentsPerItem":20,"maxCostUsd":0.5} — 10 articles ($0.003 each) + 200 comments ($0.0005 each) ≈ $0.13. The same works for blogs (blog_search + fullText) and places (place_search + reviewStats + reviews).

Naver blog search & posts

• blog_search – title, snippet, blog name/id, date, thumbnail, URL; 30 per page, paged automatically; sortBy relevance/date; dateFrom/dateTo. include: ["fullText"] opens each post.
• blog_posts – full post text, images, category, date and comment count (count only).
• blog_user_posts – a blogger's latest 50 public posts (title, summary, tags, category, date).

{"operation":"blog_search","keywords":["제주 맛집"],"maxItemsPerInput":30,"sortBy":"date"}
{"operation":"blog_posts","urls":["https://blog.naver.com/jellaby__me/224424031119"]}
{"operation":"blog_user_posts","users":["jellaby__me"],"maxItemsPerInput":20}

Naver news, ranking & comments

• news_search – title, press, time, snippet, URL, Naver article id; 10 per page; sortBy relevance/date; dateFrom/dateTo. Add include: ["fullText","comments"].
• news_ranking – most-viewed articles (5 per press, ~80 presses; rank-1 of every press first). sortBy: "comments" = most-commented. sections filters by press. dateFrom = ranking of that day.
• news_articles – full article: press, reporter, section, published/updated time, summary, body, images, original URL, comment count. Sports (m.sports) and entertainment (m.entertain) articles work too (comment count/comments not available for those).
• news_comments – comment text, likes, dislikes, reply count, time, best flag, author as Naver masks it (ab****).

{"operation":"news_search","keywords":["삼성전자"],"include":["fullText","comments","insights"],"maxItemsPerInput":10,"commentsPerItem":20}
{"operation":"news_ranking","maxItems":20,"outputMode":"lean"}
{"operation":"news_articles","urls":["https://n.news.naver.com/mnews/article/008/0005425552","https://m.sports.naver.com/article/003/0014250704"]}
{"operation":"news_comments","urls":["https://n.news.naver.com/mnews/article/022/0004166216"],"sortBy":"date","maxItemsPerInput":200}

Naver Place search, details & reviews (Naver Map)

• place_search – place id, name, category, address, road address, phone, coordinates, rating, visitor and blog review counts, booking/Naver Pay flags, amenities; 50 per page. include: ["details"] adds business hours, menu with prices, payment methods, directions; include: ["reviews"] adds reviews.
• place_details – the same detail fields by place id or URL (m.place, map.naver.com, naver.me), plus blog review count, booking flag and Naver Booking URL.
• place_reviews – visitor reviews: rating, text, visit count, visit date, voted keywords, media count, owner reply. Reviewer names and ids are never returned.
• include: ["reviewStats"] (place_search, place_details; same place price) – Naver's own review summary on each place row: average rating, review totals, top voted keywords (e.g. 음식이 맛있어요 29,610 votes, shareOfVoters 0.97), most-mentioned menu items and review themes (맛, 서비스, 분위기 …).

{"operation":"place_search","keywords":["강남역 맛집"],"maxItemsPerInput":20,"include":["details"]}
{"operation":"place_details","urls":["https://map.naver.com/p/entry/place/38562941"],"include":["reviewStats","reviews"],"reviewsPerPlace":20}
{"operation":"place_reviews","ids":["38562941"],"maxItemsPerInput":100,"deltaKey":"my-store"}

Naver KiN Q&A (지식iN)

• kin_search – question title, snippet, date, answer count, category, tags; include: ["fullText"] adds question text and answers.
• kin_questions – question text, views, date, tags and every answer (text, date, adopted flag).

{"operation":"kin_search","keywords":["노트북 추천"],"maxItemsPerInput":20}
{"operation":"kin_questions","urls":["https://kin.naver.com/qna/detail.naver?dirId=10102&docId=495288897"]}

Naver web search, autocomplete & Shopping Insight keywords

• web_search – Naver web results (SERP, the 웹 tab: sites, blogs, cafe and influencer cards): rank, title, URL, domain, site name, snippet; 20 per page, paged automatically up to the ~180 results Naver serves per keyword.
• autocomplete – up to 10 ranked Naver suggestions per keyword.
• shopping_keywords – DataLab Shopping Insight top keywords per category and period (rank 1–500), with device, gender, ages, timeUnit filters.

{"operation":"web_search","keywords":["환율"],"maxItemsPerInput":45}
{"operation":"autocomplete","keywords":["아이폰","캠핑"]}
{"operation":"shopping_keywords","categories":["50000003"],"dateFrom":"2026-10-01","dateTo":"2026-10-07","maxItemsPerInput":100}

Naver Webtoon & stock quotes

{"operation":"webtoons","maxItems":50}
{"operation":"stock_quotes","ids":["005930"],"keywords":["카카오"]}

stock_quotes returns price, change, %, open/high/low, volume, trading value, market cap, PER, EPS, PBR, dividend yield, 52-week range and foreign ownership.

For AI agents

• Call shape: operation + one array (keywords, urls, ids, users, categories). auto routes by input type. Nothing is required.
• Free first steps: {"operation":"capabilities"} returns every operation, field, include option, price and limit; "dryRun": true returns the plan and cost estimate without fetching.
• Control: maxItems (total billable rows, default 50), maxItemsPerInput, fields (projection), include, outputMode (compact · full · lean · table), maxTextChars, maxCostUsd, deltaKey.
• Errors never fail silently: each bad input gives a free row {"type":"error","errorCode":"NOT_FOUND","message":"…","fix":"…"}; codes: MISSING_INPUT, INVALID_INPUT, INVALID_URL, NOT_FOUND, NO_RESULTS, LOGIN_REQUIRED, UNSUPPORTED_OPERATION, BUDGET_REACHED, BLOCKED, UPSTREAM_ERROR. The run status message states rows, errors, cost and the first fix.
• Summary row last: rows by type, billed events, cost, stop reason, warnings, next steps, timings (also in key-value record RUN_SUMMARY), plus trust fields: perInput (requested vs returned per input, status complete · source_exhausted_or_filtered · stopped · error), isPartial, pagesFetched.
• Partial rows are flagged: when a requested enrichment (fullText, details, reviewStats) fails for a row, the row is kept with isPartial: true and unavailableFields and billed at the cheaper base event. null = unknown / not given by Naver; 0 = Naver reported zero.
• Free insights row: include: ["insights"] adds one unbilled insights row before the summary: top terms, top review keywords, date range, comment like ratio, average review rating and a lexicon-based sentiment count (labelled heuristic; not a model).
• Fair maxItems: with several inputs, every input gets at least one row of the maxItems budget before any input takes the rest.
• Fast default: an empty input returns 50 ranking rows in ~1–2 s; most single-input calls finish in 1–5 s, well inside a 30 s wait. Default memory 256 MB (benchmark runs peak at 10–40 MB).

Example lean row:

{"type":"place","id":"38562941","name":"고수닭갈비","category":"닭갈비","roadAddress":"서울 강남구 강남대로96길 12","rating":4.82,"visitorReviewCount":33454,"blogReviewCount":2026}

FAQ

Do I need a Naver account, cookies or an API key? No. Everything is public data read without login.

What does it cost? You pay per row. Examples per 1,000 rows: blog/news/KiN/web search rows $2, full posts or articles $3, places $1, reviews or comments $0.50, ranking/webtoon/stock rows $1. 100 place reviews = $0.05; 20 news articles with 20 comments each = $0.06 + $0.20. Set maxCostUsd to cap a run, or dryRun: true for a free estimate.

How many results can I get? Blog search ~1,000 per keyword, news search ~4,000, KiN ~1,000, web search ~180, Place search up to 300 places per keyword, up to 5,000 visitor reviews per place and 5,000 comments per article per run (maxItems max 5,000), Shopping Insight ranks 1–500.

Does it work with English keywords? Yes for common terms (305 curated phrases such as Gangnam restaurants, Samsung Electronics); each row shows the Korean queryUsed. Set expandQuery: false to search exactly what you typed.

How do I monitor new reviews or comments? Run on a schedule with the same deltaKey; later runs return and bill only rows not returned before (60 days memory).

What are the limits? See "Not included" below. Sports/entertainment article comments, Naver Cafe, blog comments and Naver Shopping are not covered. Review order is Naver's default (about newest first).

Not included

• Naver Cafe posts and comments (Naver's Sep 2026 notice; most Cafe content is members-only).
• Blog comments (commenter privacy; Naver blocks guest access). Blog posts include the comment count.
• Naver Shopping / SmartStore products and reviews (blocked for automated access). Use shopping_keywords for popular product keywords.
• DataLab search-trend ratios, Naver Land, image/video search and AI briefing are planned.

Requests for these return an UNSUPPORTED_OPERATION row (free) with an alternative.

Privacy

Personal data is minimised for Korea's PIPA: no reviewer or commenter ids, no profile links or avatars; place reviewer names are dropped; news comment authors appear only in the masked form Naver itself displays; KiN answerer names are dropped. Use the data in line with Naver's terms and the laws that apply to you.

Unofficial; not affiliated with or endorsed by NAVER Corp.