Naver Blog Post Search & Full Content Extractor

Naver Blog is South Korea's largest blogging platform, and its posts hold first-hand reviews, local guides, recipes, and daily-life writing that rarely appear elsewhere. This actor searches Naver Blog by keyword, walks an entire public blog archive, or fetches a list of post URLs — returning clean post records with exact publish times, full text, images, tags, categories, and places.

No Naver account, login, API key, or proxy setup is required.

What this actor does

• Three modes: search for keyword discovery, blogPosts for a blog's public archive, and postDetails for exact URLs.
• Multiple targets per run: search up to 50 keywords and read up to 50 blog IDs in one run, with duplicates removed across queries.
• Real sorting and date filters: relevance or newest-first, plus last hour/day/week/month/3 months/6 months/year and custom YYYY-MM-DD ranges.
• Full posts, not snippets: optional content enrichment adds the complete body, exact publish time, images, thumbnail, tags, category, and Naver Map places.
• Exact publish times: publishedAt is the post's real timestamp in Korea Standard Time (UTC+9), while publishDateText preserves what Naver displayed.
• Reactions and comments: like counts, reaction breakdowns, and optional comment counts and paginated comment threads (up to maxComments per post).
• Empty fields are omitted: every record is cleaned recursively, so absent data is never returned as null, "", or [].

Output per post

Field	Description
title	Post title, HTML removed
summary	Search snippet or the post page's description
authorName	Author's display name
authorUrl	Author profile link, when Naver exposes one
blogName	Name of the blog
blogId	Blog identifier, e.g. travel_lover
blogUrl	Blog home URL
postUrl	Canonical post URL
publishDate	Publish date in YYYY-MM-DD format (KST)
publishedAt	Exact publish time as ISO 8601 with +09:00 (content enrichment only)
publishDateText	Date as Naver displayed it, e.g. 2주 전 or 2026. 10. 23. 11:10
category	Blog category name (content enrichment only)
tags	Hashtags found on the post (content enrichment only)
images	Full-size content image URLs (content enrichment only)
imageCount	Number of images in images
thumbnail	Representative image for the post
places	Naver Map places mentioned in the post (content enrichment only)
hasVideo	true when the post contains a video (only present when true)
isSelfPurchased	true when Naver shows its structured self-purchased review disclosure (내돈내산)
content	Complete post text, never truncated (content enrichment only)
contentHtml	Post body HTML, only when includeHtml is on
contentLength	Character count of content
rank	1-based organic position for the query that found the post (search only)
searchQuery	First query that found the post (search only)
matchedQueries	Every processed query that found the post (search only)
commentCount	Total comments including replies; from the blog list or includeComments
comments	Comment threads up to maxComments (default 100), fetched in pages of 100, only when includeComments is on
likeCount	Main "like" (공감) count when Naver publishes it (content enrichment only)
reactionCounts	Counts by reaction type (like, impressive, thanks, wow, haha, sad) when Naver publishes them
recordType	Always "blogPost"
scrapedAt	UTC timestamp when the record was scraped

Each place in places contains name, address, latitude, longitude, phone, placeId, and bookingUrl, with empty values omitted.

With includeContent off, search and blog-archive records contain list-level fields only: title, summary, author, blog, URLs, publishDateText, and the date when Naver shows an absolute date. postDetails always fetches the post page because that is the mode's purpose.

Input

Field	Type	Default	Description
mode	string (required)	search	search, blogPosts, or postDetails
searchQueries	array of strings	["제주도 여행"]	Keywords to search (mode=search); max 50
blogIds	array of strings	–	Blog IDs or blog URLs (mode=blogPosts); max 50
postUrls	array of strings	–	Post URLs to fetch (mode=postDetails); desktop, mobile, PostView, or blogId/logNo
sortBy	string	sim	sim for relevance, date for newest first (mode=search)
publishedWithin	string	all	all, 1h, 1d, 1w, 1m, 3m, 6m, 1y, or custom (mode=search)
dateFrom	string	–	Start date YYYY-MM-DD for a custom range (mode=search)
dateTo	string	–	End date YYYY-MM-DD for a custom range (mode=search)
categoryNo	integer	0	Blog category number; 0 means every category (mode=blogPosts)
maxItems	integer	20	Hard cap on emitted records across all targets (1–10,000)
includeContent	boolean	true	Fetch each post page for full text and detail fields
includeHtml	boolean	false	Add contentHtml; turns content enrichment on
includeComments	boolean	false	Add commentCount and comment threads up to maxComments; turns content enrichment on
maxComments	integer	100	Maximum comments returned per post with includeComments (1–1,000)
maxConcurrency	integer	3	Parallel post-page requests when enriching content

Example: Korean keyword search, newest first

{
  "mode": "search",
  "searchQueries": ["제주도 여행"],
  "sortBy": "date",
  "publishedWithin": "all",
  "maxItems": 50,
  "includeContent": true
}

Example: several queries with a custom date range

{
  "mode": "search",
  "searchQueries": ["K-beauty routine", "올리브영 추천"],
  "sortBy": "sim",
  "publishedWithin": "custom",
  "dateFrom": "2026-01-01",
  "dateTo": "2026-03-31",
  "maxItems": 200,
  "includeContent": false
}

Example: a blog's full archive in one category

{
  "mode": "blogPosts",
  "blogIds": ["ranto28"],
  "categoryNo": 17,
  "maxItems": 500,
  "includeContent": true,
  "includeHtml": false
}

Example: full details for known posts

{
  "mode": "postDetails",
  "postUrls": [
    "https://blog.naver.com/e_s_l/224451062514",
    "https://m.blog.naver.com/ranto28/224461024209",
    "https://blog.naver.com/PostView.naver?blogId=hoyaaaaa-&logNo=224458179821"
  ]
}

Search limits and blog archives

Naver returns at most 1,000 posts per keyword search. This is a source-side ceiling, not an actor setting. To collect more:

• split a broad topic into narrower queries;
• use publishedWithin presets or a custom dateFrom/dateTo window and run several windows;
• deduplicate with postUrl — the actor already removes duplicates inside a run.

Blog archives use a different source and are not limited to 1,000 posts: a public blog with 2,700 posts can return all 2,700. Search results are likely to repeat across pages, so deduplication is applied by canonical post URL.

Comment counts are included when Naver exposes them: from the blog list in blogPosts mode, or from the post page when includeComments is on. Likes (likeCount and reactionCounts) come from Naver's public reaction service and are omitted for posts with no reactions. With includeComments, comments returns up to maxComments per post (default 100, fetched in pages of 100) including replies when Naver returns them; commentCount is always the full total, so a post can have more comments than rows in comments.

Limitations

• Search ceiling: Naver accepts search start positions up to 1,000, so one query exposes roughly a 1,000-result window; after deduplication a run can return fewer. Split by date or query for more.
• Comment cap: Naver serves comment pages cookielessly; the actor pages through them up to maxComments per post (default 100, max 1,000) to keep output and request cost bounded. commentCount still reports the full total. Increase the cap if you need more rows. Posts with comments disabled return neither field.
• Not shipped (scope decisions, not source limits): blogger profile totals (neighbors, visitors, post count), cafe/news article search, paid-sponsorship classification, and incremental only-new state. Profile and cafe/news data are reachable on separate Naver surfaces but are outside this post-oriented actor; paid sponsorship has no ground-truth label and remains a text heuristic for users; incremental state is an implementation choice, and date windows plus postUrl dedupe cover the same workflow. Self-purchased reviews are labeled by Naver and are exposed as isSelfPurchased.
• Private, deleted, or members-only posts are skipped; one bad post does not fail the run.
• Search relevance is Naver's own ranking, including Korean results for English queries; the actor does not re-rank.
• Markup changes: Naver redesigns can require a parser update. The actor ships fixtures and status messages instead of failing silently.

Use cases

• Consumer research — collect how Korean bloggers describe a product, brand, or store before a launch.
• Influencer and blog discovery — find active blogs in K-beauty, food, travel, parenting, or finance.
• Brand monitoring — track mentions across the Korean blogosphere, then filter by exact publish date.
• Content and trend analysis — measure post volume and topic shifts across date windows.
• Local travel research — gather first-hand restaurant, cafe, and itinerary write-ups with mapped places.
• Archive and migration — export a public blog's full post history, including text and image URLs.

FAQ

Do I need a Naver account or API key? No. The actor reads public Naver Blog pages. There is no login, key, or proxy configuration to provide.

Can I search in Korean and English? Yes. Korean, English, and mixed queries all work. Results are mostly Korean-language blogs because that is what Naver Blog contains.

How do I find a blog ID? Use the segment right after blog.naver.com/. For https://blog.naver.com/travel_lover, the blog ID is travel_lover. The actor also accepts the full blog URL or @travel_lover.

Why do some posts have no content, images, or publishedAt? Those fields come from the post page, so they require includeContent. They are also omitted for empty-body posts, private or deleted posts, and text-only posts with no images or places.

What is the difference between publishDate and publishedAt? publishDate is the date in YYYY-MM-DD format. publishedAt is the exact timestamp with a +09:00 offset. publishDateText keeps the original wording, which may be relative such as 3일 전.

Why did a keyword return fewer posts than expected? Naver caps one keyword at 1,000 results and may return fewer for narrow queries. Add more queries or split the period into smaller date windows.

Why are there no likes or comments on some posts? Naver's reaction and comment services are public but conditional: posts with no reactions omit likeCount/reactionCounts, and posts with comments disabled omit commentCount/comments. Set includeComments=true to request comments; commentCount is the total and comments holds up to maxComments rows (default 100, paginated 100 at a time).

What does isSelfPurchased mean? It is true only when Naver itself shows the structured self-purchased review disclosure (내돈내산) on the post. It is not a paid-sponsorship classifier; 협찬/광고 wording remains in content for your own rules.

What happens to private, deleted, or blocked posts? They are skipped with a log message. One missing post does not fail the run. If every target is invalid, the run finishes with a status message explaining what to check.

Can I fetch more than 10,000 posts? maxItems is capped at 10,000 per run. For larger archives, split the work into separate runs and deduplicate by postUrl.

Is this affiliated with Naver? No. This is an independent third-party actor that reads public Naver Blog pages. It is not affiliated with, endorsed by, or sponsored by Naver Corporation.