# SocialCrawl API: naver endpoints
# Base URL: https://www.socialcrawl.dev
# Auth: x-api-key header
# Full docs: https://www.socialcrawl.dev/docs/naver

## GET /v1/naver/blog/search
Search Naver Blog
Credit cost: 1-9 (metered): 1 credit per page. Relevance against your query (by default, or with relevance=) is free. Adding relevant_to= holds 4 extra credits and refunds down to 1 credit per started 25 rows that were newly judged: rows already judged for the same topic are free, a page where nothing could be judged refunds the whole extra, and a cached page is free. The default labels (sponsored, intent, niche) are free. label=mention, quality, ai_style, ai_written, injection, or label=intent with offer=, holds 4 extra credits and refunds down to 1 credit per started 25 posts that were newly judged: posts already labelled are free, a page where nothing could be judged refunds the whole extra, and a cached page is free. With `max_pages`, `max_items` or `max_credits`, each page walked is billed at this price (a cached page is free), and `max_credits` caps the total.
Parameters:
- query (required): Free-text search term (UTF-8). Required. Search short noun phrases (for example `다한증` or `다한증 수술`), which match best: Naver does not treat quotes as an exact phrase. Add `relevance=filter` to drop the rows that are not about the query.
- display (optional, integer): Number of items to return per page. Defaults to 10, cap 100; `display=101` returns 400.
- start (optional, integer): 1-indexed offset for pagination. Defaults to 1, cap 1000; `start=1001` is a 400 that costs nothing, and the page that reaches rank 1000 is the last one (`has_more: false`). Past rank 100, `sim` and `point` pages repeat Naver's top 100: those rows are removed and counted in `_warnings` (`recycled_rows`), and an emptied page is free. Page deeper with `sort=date`.
- sort (optional, enum: sim | date): Sort order. Accepted values: sim|date. Defaults to `sim` (relevance) when the corpus supports sort. Values outside this corpus's domain are rejected with a 400 before any upstream call. `date` lists the newest first, but Naver serves only the first 1,000 rows of a query (`start` caps at 1000), so on a common word it reaches back hours or days, not months, and on a multi-word query Naver also loosens the match. To cover a longer window, search narrower terms so the newest 1,000 rows span it, and add `since=`: rows dated before it are left off each page.
- relevance (optional, enum: score | filter): Optional. Without this param every row already carries computed.relevance against your query, free ({ p, sense, depth, spam }: is this row about what your query means, or a different thing that shares its words?), and nothing is dropped or reordered. score asks for it explicitly and waits for every row; filter also drops the rows that are not about your query, and lists their ids in data.relevance.dropped_ids. Your query is the topic; nothing to configure. A row that could not be judged is never dropped and carries relevance: null. Pagination is unchanged, so a filtered page can hold fewer rows. Free with your query as the topic; with relevant_to it adds 1 credit per started 25 newly judged rows (rows already judged for the same topic are free, and so is a cached page).
- relevance_threshold (optional, string): Optional, only with relevance. The probability (0 to 1) a row must reach to be kept by relevance=filter. Default 0.5. Lower keeps more rows, higher keeps fewer.
- relevant_to (optional, string): Optional, only with relevance. Up to 200 characters describing what you mean, used as the topic instead of the query. Use it when the query is ambiguous, for example query=cleopatra with relevant_to=Cleopatra, the IGT slot game.
- judgments (optional, enum: on | off): Optional, on (the default) or off. By default every row gains free SocialCrawl judgments (computed.labels, and computed.relevance on search endpoints), reported in data.labels (mode default) and data.relevance (origin default), each with a status (complete, partial or skipped) and pending: the rows still being judged when the page was sent, which carry null now and are filled on your next call or cached read. Default judgments never add credits, never change an existing field, and never drop or reorder a row. off returns the page exactly as before, with none of those keys. label=none does the same.
- dry_run (optional, enum: 1): Optional. When 1, return a cost preview for this labelled or relevance-filtered request without fetching the page or judging any row. data.estimate reports rows_expected, rows_cached, label_credits_min, label_credits_max and base_credits. 0 credits charged.
- fit (optional, enum: goal): Optional. When goal, keep the rows and fields needed for the goal you pass in goal= (plus any that are uncertain, and the first and last), and replace the rest with a stub. data.held_back lists the held ids and a recall id that re-reads the full page from cache at no extra charge. Without this param the page is unchanged.
- goal (optional, string): Required by fit=goal. What you are trying to do, in your own words, up to 300 characters.
- fit_tokens (optional, integer): Optional, only with fit=goal. Soft cap on how much of the page to keep, in tokens. Uncertain blocks and the first and last block are kept even if they exceed it.
- label (optional, string): Optional CSV of SocialCrawl labels to add to every post. Without this param every page already carries sponsored, intent and niche, free; label= adds the labels you name to them (the defaults keep running). sponsored, intent and niche are free when asked for too; mention, quality, ai_style, intent with offer=, and injection add 1 credit per started 25 newly judged posts. judgments=off (or label=none) turns the default labels off. sponsored: is the post a paid or gifted promotion (p, 0 to 1), did it carry a disclosure marker such as #ad or 광고 (disclosed), is it likely paid with no marker (undisclosed), and which of the accounts it mentions does it promote (brand, or null). These are signals to review, never a finding. mention (needs brand=): is the post about that brand rather than something that shares its name (about_brand, 0 to 1), how it feels about the brand on five levels (sentiment_level 0 to 4 and sentiment_score 0 to 1, null when the post is not about the brand), is it sarcastic, which aspect it talks about (taste_or_quality, price_or_value, availability_or_delivery, health_or_safety, advertising_or_campaign, customer_service, none), and did the author buy or use it (first_hand). intent: what the author is mainly doing (label: asking_for_recommendation, comparing_options, switching_away, complaining, promoting, news_or_discussion, other, or null when unsure, with confidence), whether they read as a potential buyer rather than a seller (buyer, seller), how pressing the need is (urgency, 0 to 3), and, when you pass offer=, whether your offer would plausibly help them (fits_offer). niche: which of the 33 niches of the published taxonomy sc-niche-v1 the caption belongs to, or personal_no_niche, or other (label, confidence, taxonomy), with label null when the caption is too thin to tell or the pick is unsure. It reads the caption only, not the video. quality: how much checkable detail the caption carries (fact_density 0 to 3), whether it mainly asks for likes, replies, shares, follows or tags (engagement_bait, 0 to 1), whether it is written to provoke anger as a way to get engagement (rage_bait, 0 to 1; about the writing, never the side taken), whether it only repeats someone else's news or view (secondhand, 0 to 1), and what the post is mainly doing (post_aim: inform, opinion, sell, entertain, provoke, other). exclude=engagement_bait drops posts whose engagement_bait is 0.8 or higher. ai_style: writing patterns common in AI-generated text, judged on the text alone. It does not say who wrote the post, and a person who writes in this style is flagged too. Six patterns are judged, each a probability (0 to 1): binary_contrast ("This is not X. It's Y."), faux_insight ("Here's what nobody tells you" and throat-clearing openers), colon_reveal ("The best part: it learns."), profound_kicker (ends on an aphorism or mic-drop line), puffery ("marks a pivotal moment", "highlighting the importance of") and rhetorical_setup ("What if I told you", a question the post answers itself, "Not a X. Not a Y. A Z."). Four are counted in the text: slop_words (words and phrases such as delve, leverage, robust, game changer, at the end of the day), em_dashes, emoji_bullets (lines that start with an emoji or a symbol bullet) and one_line_paragraphs (paragraphs that are one short line). pattern_count is how many patterns are present (a judged pattern at 0.8 or higher, slop_words 1 or more, em_dashes 2 or more, emoji_bullets 2 or more, one_line_paragraphs 5 or more) and markers lists their names, comma-separated. A post with less than about eight words of text is only counted, with every judged pattern 0. The judged patterns read the first 400 and the last 150 characters of a long post, and the counts read the whole post. exclude=ai_style drops posts with 2 or more patterns. ai_written: how likely the post was written by an AI model, from SocialCrawl's own text classifier, reading the opening of the post (about 1,000 characters): p (0 to 1, calibrated), level (likely (p 0.95 or higher), possible (p 0.71 or higher) or unlikely) and model (the model name; a new name means a new scale). A post under 8 words is too short to read: p null, level too_short. It is a probability, not proof: a person can be flagged and an AI-written post can pass, and the shorter the post the less certain it is. exclude=ai_written drops likely posts. injection flags text that addresses an AI system and tries to direct it (flagged, p); it never drops or rewrites a row. A post that could not be judged carries labels: null. Posts already labelled for anyone are free, and so is a cached page. data.labels reports what was judged and billed.
- exclude (optional, string): Optional CSV (engagement_bait, ai_style, ai_written). engagement_bait, only with label=quality, drops posts whose engagement_bait probability is 0.8 or higher. ai_style, only with label=ai_style, drops posts that show 2 or more of its writing patterns (pattern_count 2 or higher). ai_written, only with label=ai_written, drops posts whose level is likely (p 0.95 or higher), or whose p reaches your ai_written_threshold; a post too short to read is never dropped. Dropped posts are listed in data.labels.dropped_ids. A post that could not be judged is never dropped.
- ai_written_threshold (optional, string): Optional, only with label=ai_written and exclude=ai_written. The p (0 to 1) at which exclude=ai_written drops a post. Default 0.95 (level likely). 0.71 also drops possible posts: more AI posts go, and more posts by people with them.
- brand (optional, string): Required by label=mention, ignored otherwise. The brand or product the posts are judged against, up to 300 characters, for example brand=Buldak. Without it label=mention is skipped with the warning label_mention_needs_brand and is not billed.
- brand_description (optional, string): Optional, with label=mention. One plain sentence saying what the brand is, for example brand_description=Samyang's spicy instant noodle brand. Use it when the name is also an ordinary word or another company's name.
- offer (optional, string): Optional, with label=intent. One or two plain sentences saying what you sell, up to 300 characters, for example offer=A web design agency that builds websites for small businesses. Adds fits_offer to every post; without it fits_offer is null.
- label_evidence (optional, enum: 1): Optional, only with label=. When 1, every labelled row also carries computed.labels_evidence.<preset> = { quote, sentence_index }: the sentence in the row that most clearly shows the label, copied verbatim. Absent or null when no single sentence shows it.
- since (optional, string): Only rows published on or after this date: YYYY-MM-DD (midnight UTC) or an ISO 8601 timestamp. Rows older than it are left off each page. This list is not in date order, so the walk does not stop at the date: keep following `next_cursor`, and each page costs what a page costs even when every row on it is filtered out. A row with no date is kept. `published_after`, `start_date`, `from_date`, `publish_time` (`yesterday`, `this-week`, `this-month`, `last-3-months`, `last-6-months`) and `max_age_days` are accepted and applied as `since`.
- until (optional, string): Only rows published on or before this date: YYYY-MM-DD (that whole UTC day is included) or an ISO 8601 timestamp. Newer rows are left off each page and the walk does not stop, so older pages follow. A row with no date is kept. `published_before`, `end_date` and `to_date` are accepted and applied as `until`.
- max_pages (optional, integer): 1 to 50 (default 1, or 50 when `max_items` or `max_credits` is sent). Walk up to this many pages in one call and return the rows from all of them. Each page walked is billed exactly as one call to this endpoint (a cached page is free) and counts against your rate limit; its charge and any refund are recorded under the page's own request id, listed in `data.walk.request_ids`. The walk stops early at the last page, or when the request's time budget runs low; `data.walk.stopped` says why (`end`, `max_pages`, `max_items`, `max_credits`, `time_budget`, `page_error`) and `pagination.next_cursor` continues from where it stopped.
- max_items (optional, integer): 1 to 5000. Get more in one call: keep walking pages until the response holds at least this many rows, then stop with `data.walk.stopped: "max_items"`. The last page is returned whole (it was billed), so `data.items` can hold up to one page more than this, and `pagination.next_cursor` resumes right after it without skipping a row. Walks up to 50 pages unless `max_pages` says fewer; each page is billed as one call (a cached page is free). Add `max_credits` to cap the spend. A long walk can stop on the time budget first (`"time_budget"`): send the returned cursor with the same params to continue.
- max_credits (optional, integer): 1 to 100000. The most this call may spend, across every page it walks. A page starts only while the credits already spent plus that page's price (or, when higher, the most any page of this walk has cost) still fit, so the call never goes past it; the walk then stops with `data.walk.stopped: "max_credits"` and the cursor to resume. Sent alone, it walks as far as the budget goes (up to 50 pages). A value below the price of one page is a free 400 whose `error.details.page_credits` names that price. Credits a row join (`include=`) adds are known only once a page has run, so page 1 is checked against its own price.
Worth sending:
- label=sponsored: Marks each post that is paid or sponsored, at no extra credits; intent and niche are free too.
- relevance=filter: Drops rows that are not about your query, at no extra credits; relevance=score keeps them and adds a score.
curl -G "https://www.socialcrawl.dev/v1/naver/blog/search" \
  --data-urlencode "query=소셜크롤" \
  --data-urlencode "label=sponsored" \
  --data-urlencode "relevance=filter" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/news/search
Search Naver News
Credit cost: 1-9 (metered): 1 credit per page. Relevance against your query (by default, or with relevance=) is free. Adding relevant_to= holds 4 extra credits and refunds down to 1 credit per started 25 rows that were newly judged: rows already judged for the same topic are free, a page where nothing could be judged refunds the whole extra, and a cached page is free. The default labels (sponsored, intent, niche) are free. label=mention, quality, ai_style, ai_written, injection, or label=intent with offer=, holds 4 extra credits and refunds down to 1 credit per started 25 posts that were newly judged: posts already labelled are free, a page where nothing could be judged refunds the whole extra, and a cached page is free. With `max_pages`, `max_items` or `max_credits`, each page walked is billed at this price (a cached page is free), and `max_credits` caps the total.
Parameters:
- query (required): Free-text search term (UTF-8). Required. Search short noun phrases (for example `다한증` or `다한증 수술`), which match best: Naver does not treat quotes as an exact phrase. Add `relevance=filter` to drop the rows that are not about the query.
- display (optional, integer): Number of items to return per page. Defaults to 10, cap 100; `display=101` returns 400.
- start (optional, integer): 1-indexed offset for pagination. Defaults to 1, cap 1000; `start=1001` is a 400 that costs nothing, and the page that reaches rank 1000 is the last one (`has_more: false`). Past rank 100, `sim` and `point` pages repeat Naver's top 100: those rows are removed and counted in `_warnings` (`recycled_rows`), and an emptied page is free. Page deeper with `sort=date`.
- sort (optional, enum: sim | date): Sort order. Accepted values: sim|date. Defaults to `sim` (relevance) when the corpus supports sort. Values outside this corpus's domain are rejected with a 400 before any upstream call. `date` lists the newest first, but Naver serves only the first 1,000 rows of a query (`start` caps at 1000), so on a common word it reaches back hours or days, not months, and on a multi-word query Naver also loosens the match. `since=` sets `sort=date` and ends the walk at the date, so narrow the query to reach further back.
- relevance (optional, enum: score | filter): Optional. Without this param every row already carries computed.relevance against your query, free ({ p, sense, depth, spam }: is this row about what your query means, or a different thing that shares its words?), and nothing is dropped or reordered. score asks for it explicitly and waits for every row; filter also drops the rows that are not about your query, and lists their ids in data.relevance.dropped_ids. Your query is the topic; nothing to configure. A row that could not be judged is never dropped and carries relevance: null. Pagination is unchanged, so a filtered page can hold fewer rows. Free with your query as the topic; with relevant_to it adds 1 credit per started 25 newly judged rows (rows already judged for the same topic are free, and so is a cached page).
- relevance_threshold (optional, string): Optional, only with relevance. The probability (0 to 1) a row must reach to be kept by relevance=filter. Default 0.5. Lower keeps more rows, higher keeps fewer.
- relevant_to (optional, string): Optional, only with relevance. Up to 200 characters describing what you mean, used as the topic instead of the query. Use it when the query is ambiguous, for example query=cleopatra with relevant_to=Cleopatra, the IGT slot game.
- judgments (optional, enum: on | off): Optional, on (the default) or off. By default every row gains free SocialCrawl judgments (computed.labels, and computed.relevance on search endpoints), reported in data.labels (mode default) and data.relevance (origin default), each with a status (complete, partial or skipped) and pending: the rows still being judged when the page was sent, which carry null now and are filled on your next call or cached read. Default judgments never add credits, never change an existing field, and never drop or reorder a row. off returns the page exactly as before, with none of those keys. label=none does the same.
- dry_run (optional, enum: 1): Optional. When 1, return a cost preview for this labelled or relevance-filtered request without fetching the page or judging any row. data.estimate reports rows_expected, rows_cached, label_credits_min, label_credits_max and base_credits. 0 credits charged.
- fit (optional, enum: goal): Optional. When goal, keep the rows and fields needed for the goal you pass in goal= (plus any that are uncertain, and the first and last), and replace the rest with a stub. data.held_back lists the held ids and a recall id that re-reads the full page from cache at no extra charge. Without this param the page is unchanged.
- goal (optional, string): Required by fit=goal. What you are trying to do, in your own words, up to 300 characters.
- fit_tokens (optional, integer): Optional, only with fit=goal. Soft cap on how much of the page to keep, in tokens. Uncertain blocks and the first and last block are kept even if they exceed it.
- label (optional, string): Optional CSV of SocialCrawl labels to add to every post. Without this param every page already carries sponsored, intent and niche, free; label= adds the labels you name to them (the defaults keep running). sponsored, intent and niche are free when asked for too; mention, quality, ai_style, intent with offer=, and injection add 1 credit per started 25 newly judged posts. judgments=off (or label=none) turns the default labels off. sponsored: is the post a paid or gifted promotion (p, 0 to 1), did it carry a disclosure marker such as #ad or 광고 (disclosed), is it likely paid with no marker (undisclosed), and which of the accounts it mentions does it promote (brand, or null). These are signals to review, never a finding. mention (needs brand=): is the post about that brand rather than something that shares its name (about_brand, 0 to 1), how it feels about the brand on five levels (sentiment_level 0 to 4 and sentiment_score 0 to 1, null when the post is not about the brand), is it sarcastic, which aspect it talks about (taste_or_quality, price_or_value, availability_or_delivery, health_or_safety, advertising_or_campaign, customer_service, none), and did the author buy or use it (first_hand). intent: what the author is mainly doing (label: asking_for_recommendation, comparing_options, switching_away, complaining, promoting, news_or_discussion, other, or null when unsure, with confidence), whether they read as a potential buyer rather than a seller (buyer, seller), how pressing the need is (urgency, 0 to 3), and, when you pass offer=, whether your offer would plausibly help them (fits_offer). niche: which of the 33 niches of the published taxonomy sc-niche-v1 the caption belongs to, or personal_no_niche, or other (label, confidence, taxonomy), with label null when the caption is too thin to tell or the pick is unsure. It reads the caption only, not the video. quality: how much checkable detail the caption carries (fact_density 0 to 3), whether it mainly asks for likes, replies, shares, follows or tags (engagement_bait, 0 to 1), whether it is written to provoke anger as a way to get engagement (rage_bait, 0 to 1; about the writing, never the side taken), whether it only repeats someone else's news or view (secondhand, 0 to 1), and what the post is mainly doing (post_aim: inform, opinion, sell, entertain, provoke, other). exclude=engagement_bait drops posts whose engagement_bait is 0.8 or higher. ai_style: writing patterns common in AI-generated text, judged on the text alone. It does not say who wrote the post, and a person who writes in this style is flagged too. Six patterns are judged, each a probability (0 to 1): binary_contrast ("This is not X. It's Y."), faux_insight ("Here's what nobody tells you" and throat-clearing openers), colon_reveal ("The best part: it learns."), profound_kicker (ends on an aphorism or mic-drop line), puffery ("marks a pivotal moment", "highlighting the importance of") and rhetorical_setup ("What if I told you", a question the post answers itself, "Not a X. Not a Y. A Z."). Four are counted in the text: slop_words (words and phrases such as delve, leverage, robust, game changer, at the end of the day), em_dashes, emoji_bullets (lines that start with an emoji or a symbol bullet) and one_line_paragraphs (paragraphs that are one short line). pattern_count is how many patterns are present (a judged pattern at 0.8 or higher, slop_words 1 or more, em_dashes 2 or more, emoji_bullets 2 or more, one_line_paragraphs 5 or more) and markers lists their names, comma-separated. A post with less than about eight words of text is only counted, with every judged pattern 0. The judged patterns read the first 400 and the last 150 characters of a long post, and the counts read the whole post. exclude=ai_style drops posts with 2 or more patterns. ai_written: how likely the post was written by an AI model, from SocialCrawl's own text classifier, reading the opening of the post (about 1,000 characters): p (0 to 1, calibrated), level (likely (p 0.95 or higher), possible (p 0.71 or higher) or unlikely) and model (the model name; a new name means a new scale). A post under 8 words is too short to read: p null, level too_short. It is a probability, not proof: a person can be flagged and an AI-written post can pass, and the shorter the post the less certain it is. exclude=ai_written drops likely posts. injection flags text that addresses an AI system and tries to direct it (flagged, p); it never drops or rewrites a row. A post that could not be judged carries labels: null. Posts already labelled for anyone are free, and so is a cached page. data.labels reports what was judged and billed.
- exclude (optional, string): Optional CSV (engagement_bait, ai_style, ai_written). engagement_bait, only with label=quality, drops posts whose engagement_bait probability is 0.8 or higher. ai_style, only with label=ai_style, drops posts that show 2 or more of its writing patterns (pattern_count 2 or higher). ai_written, only with label=ai_written, drops posts whose level is likely (p 0.95 or higher), or whose p reaches your ai_written_threshold; a post too short to read is never dropped. Dropped posts are listed in data.labels.dropped_ids. A post that could not be judged is never dropped.
- ai_written_threshold (optional, string): Optional, only with label=ai_written and exclude=ai_written. The p (0 to 1) at which exclude=ai_written drops a post. Default 0.95 (level likely). 0.71 also drops possible posts: more AI posts go, and more posts by people with them.
- brand (optional, string): Required by label=mention, ignored otherwise. The brand or product the posts are judged against, up to 300 characters, for example brand=Buldak. Without it label=mention is skipped with the warning label_mention_needs_brand and is not billed.
- brand_description (optional, string): Optional, with label=mention. One plain sentence saying what the brand is, for example brand_description=Samyang's spicy instant noodle brand. Use it when the name is also an ordinary word or another company's name.
- offer (optional, string): Optional, with label=intent. One or two plain sentences saying what you sell, up to 300 characters, for example offer=A web design agency that builds websites for small businesses. Adds fits_offer to every post; without it fits_offer is null.
- label_evidence (optional, enum: 1): Optional, only with label=. When 1, every labelled row also carries computed.labels_evidence.<preset> = { quote, sentence_index }: the sentence in the row that most clearly shows the label, copied verbatim. Absent or null when no single sentence shows it.
- since (optional, string): Only rows published on or after this date: YYYY-MM-DD (midnight UTC) or an ISO 8601 timestamp. Rows older than it are left off the page, and the page that reaches one ends the walk: `next_cursor` is not returned and `pagination.stopped_at` is `since`. Sending it sets the newest-first sort shown in this endpoint's sort parameter (any other sort value is a free 400). The page still costs what a page costs, so you pay for the pages you walk and no more. A row with no date is kept. `published_after`, `start_date`, `from_date`, `publish_time` (`yesterday`, `this-week`, `this-month`, `last-3-months`, `last-6-months`) and `max_age_days` are accepted and applied as `since`.
- until (optional, string): Only rows published on or before this date: YYYY-MM-DD (that whole UTC day is included) or an ISO 8601 timestamp. Newer rows are left off each page and the walk does not stop, so older pages follow. A row with no date is kept. `published_before`, `end_date` and `to_date` are accepted and applied as `until`.
- max_pages (optional, integer): 1 to 50 (default 1, or 50 when `max_items` or `max_credits` is sent). Walk up to this many pages in one call and return the rows from all of them. Each page walked is billed exactly as one call to this endpoint (a cached page is free) and counts against your rate limit; its charge and any refund are recorded under the page's own request id, listed in `data.walk.request_ids`. The walk stops early at the last page, or when the request's time budget runs low; `data.walk.stopped` says why (`end`, `max_pages`, `max_items`, `max_credits`, `time_budget`, `page_error`) and `pagination.next_cursor` continues from where it stopped.
- max_items (optional, integer): 1 to 5000. Get more in one call: keep walking pages until the response holds at least this many rows, then stop with `data.walk.stopped: "max_items"`. The last page is returned whole (it was billed), so `data.items` can hold up to one page more than this, and `pagination.next_cursor` resumes right after it without skipping a row. Walks up to 50 pages unless `max_pages` says fewer; each page is billed as one call (a cached page is free). Add `max_credits` to cap the spend. A long walk can stop on the time budget first (`"time_budget"`): send the returned cursor with the same params to continue.
- max_credits (optional, integer): 1 to 100000. The most this call may spend, across every page it walks. A page starts only while the credits already spent plus that page's price (or, when higher, the most any page of this walk has cost) still fit, so the call never goes past it; the walk then stops with `data.walk.stopped: "max_credits"` and the cursor to resume. Sent alone, it walks as far as the budget goes (up to 50 pages). A value below the price of one page is a free 400 whose `error.details.page_credits` names that price. Credits a row join (`include=`) adds are known only once a page has run, so page 1 is checked against its own price.
Worth sending:
- label=sponsored: Marks each post that is paid or sponsored, at no extra credits; intent and niche are free too.
- relevance=filter: Drops rows that are not about your query, at no extra credits; relevance=score keeps them and adds a score.
curl -G "https://www.socialcrawl.dev/v1/naver/news/search" \
  --data-urlencode "query=삼성전자" \
  --data-urlencode "label=sponsored" \
  --data-urlencode "relevance=filter" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/encyc/search
Search Naver Encyclopedia
Credit cost: 1 (standard)
Parameters:
- query (required): Free-text search term (UTF-8). Required.
- display (optional, integer): Number of items to return per page. Defaults to 10, cap 100; `display=101` returns 400.
- start (optional, integer): 1-indexed offset for pagination. Defaults to 1, cap 1000; `start=1001` is a 400 that costs nothing, and the page that reaches rank 1000 is the last one (`has_more: false`). Past rank 100, `sim` and `point` pages repeat Naver's top 100: those rows are removed and counted in `_warnings` (`recycled_rows`), and an emptied page is free. This corpus ignores `sort`, so `sort=date` does not avoid the repeats.
- sort (optional, string): Sort order. Accepted values: (sort ignored). Defaults to `sim` (relevance) when the corpus supports sort. Values outside this corpus's domain are rejected with a 400 before any upstream call.
- max_pages (optional, integer): 1 to 50 (default 1, or 50 when `max_items` or `max_credits` is sent). Walk up to this many pages in one call and return the rows from all of them. Each page walked is billed exactly as one call to this endpoint (a cached page is free) and counts against your rate limit; its charge and any refund are recorded under the page's own request id, listed in `data.walk.request_ids`. The walk stops early at the last page, or when the request's time budget runs low; `data.walk.stopped` says why (`end`, `max_pages`, `max_items`, `max_credits`, `time_budget`, `page_error`) and `pagination.next_cursor` continues from where it stopped.
- max_items (optional, integer): 1 to 5000. Get more in one call: keep walking pages until the response holds at least this many rows, then stop with `data.walk.stopped: "max_items"`. The last page is returned whole (it was billed), so `data.items` can hold up to one page more than this, and `pagination.next_cursor` resumes right after it without skipping a row. Walks up to 50 pages unless `max_pages` says fewer; each page is billed as one call (a cached page is free). Add `max_credits` to cap the spend. A long walk can stop on the time budget first (`"time_budget"`): send the returned cursor with the same params to continue.
- max_credits (optional, integer): 1 to 100000. The most this call may spend, across every page it walks. A page starts only while the credits already spent plus that page's price (or, when higher, the most any page of this walk has cost) still fit, so the call never goes past it; the walk then stops with `data.walk.stopped: "max_credits"` and the cursor to resume. Sent alone, it walks as far as the budget goes (up to 50 pages). A value below the price of one page is a free 400 whose `error.details.page_credits` names that price. Credits a row join (`include=`) adds are known only once a page has run, so page 1 is checked against its own price.
curl -G "https://www.socialcrawl.dev/v1/naver/encyc/search" \
  --data-urlencode "query=양자역학" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/cafearticle/search
Search Naver Cafe articles
Credit cost: 1-9 (metered): 1 credit per page. Relevance against your query (by default, or with relevance=) is free. Adding relevant_to= holds 4 extra credits and refunds down to 1 credit per started 25 rows that were newly judged: rows already judged for the same topic are free, a page where nothing could be judged refunds the whole extra, and a cached page is free. The default labels (sponsored, intent, niche) are free. label=mention, quality, ai_style, ai_written, injection, or label=intent with offer=, holds 4 extra credits and refunds down to 1 credit per started 25 posts that were newly judged: posts already labelled are free, a page where nothing could be judged refunds the whole extra, and a cached page is free. With `max_pages`, `max_items` or `max_credits`, each page walked is billed at this price (a cached page is free), and `max_credits` caps the total.
Parameters:
- query (required): Free-text search term (UTF-8). Required. Search short noun phrases (for example `다한증` or `다한증 수술`), which match best: Naver does not treat quotes as an exact phrase. Add `relevance=filter` to drop the rows that are not about the query.
- display (optional, integer): Number of items to return per page. Defaults to 10, cap 100; `display=101` returns 400.
- start (optional, integer): 1-indexed offset for pagination. Defaults to 1, cap 1000; `start=1001` is a 400 that costs nothing, and the page that reaches rank 1000 is the last one (`has_more: false`). Past rank 100, `sim` and `point` pages repeat Naver's top 100: those rows are removed and counted in `_warnings` (`recycled_rows`), and an emptied page is free. Page deeper with `sort=date`.
- sort (optional, enum: sim | date): Sort order. Accepted values: sim|date. Defaults to `sim` (relevance) when the corpus supports sort. Values outside this corpus's domain are rejected with a 400 before any upstream call. `date` lists the newest first, but Naver serves only the first 1,000 rows of a query (`start` caps at 1000), so on a common word it reaches back hours or days, not months, and on a multi-word query Naver also loosens the match. Cafe rows carry no date, so `since=` cannot narrow them.
- relevance (optional, enum: score | filter): Optional. Without this param every row already carries computed.relevance against your query, free ({ p, sense, depth, spam }: is this row about what your query means, or a different thing that shares its words?), and nothing is dropped or reordered. score asks for it explicitly and waits for every row; filter also drops the rows that are not about your query, and lists their ids in data.relevance.dropped_ids. Your query is the topic; nothing to configure. A row that could not be judged is never dropped and carries relevance: null. Pagination is unchanged, so a filtered page can hold fewer rows. Free with your query as the topic; with relevant_to it adds 1 credit per started 25 newly judged rows (rows already judged for the same topic are free, and so is a cached page).
- relevance_threshold (optional, string): Optional, only with relevance. The probability (0 to 1) a row must reach to be kept by relevance=filter. Default 0.5. Lower keeps more rows, higher keeps fewer.
- relevant_to (optional, string): Optional, only with relevance. Up to 200 characters describing what you mean, used as the topic instead of the query. Use it when the query is ambiguous, for example query=cleopatra with relevant_to=Cleopatra, the IGT slot game.
- judgments (optional, enum: on | off): Optional, on (the default) or off. By default every row gains free SocialCrawl judgments (computed.labels, and computed.relevance on search endpoints), reported in data.labels (mode default) and data.relevance (origin default), each with a status (complete, partial or skipped) and pending: the rows still being judged when the page was sent, which carry null now and are filled on your next call or cached read. Default judgments never add credits, never change an existing field, and never drop or reorder a row. off returns the page exactly as before, with none of those keys. label=none does the same.
- dry_run (optional, enum: 1): Optional. When 1, return a cost preview for this labelled or relevance-filtered request without fetching the page or judging any row. data.estimate reports rows_expected, rows_cached, label_credits_min, label_credits_max and base_credits. 0 credits charged.
- fit (optional, enum: goal): Optional. When goal, keep the rows and fields needed for the goal you pass in goal= (plus any that are uncertain, and the first and last), and replace the rest with a stub. data.held_back lists the held ids and a recall id that re-reads the full page from cache at no extra charge. Without this param the page is unchanged.
- goal (optional, string): Required by fit=goal. What you are trying to do, in your own words, up to 300 characters.
- fit_tokens (optional, integer): Optional, only with fit=goal. Soft cap on how much of the page to keep, in tokens. Uncertain blocks and the first and last block are kept even if they exceed it.
- label (optional, string): Optional CSV of SocialCrawl labels to add to every post. Without this param every page already carries sponsored, intent and niche, free; label= adds the labels you name to them (the defaults keep running). sponsored, intent and niche are free when asked for too; mention, quality, ai_style, intent with offer=, and injection add 1 credit per started 25 newly judged posts. judgments=off (or label=none) turns the default labels off. sponsored: is the post a paid or gifted promotion (p, 0 to 1), did it carry a disclosure marker such as #ad or 광고 (disclosed), is it likely paid with no marker (undisclosed), and which of the accounts it mentions does it promote (brand, or null). These are signals to review, never a finding. mention (needs brand=): is the post about that brand rather than something that shares its name (about_brand, 0 to 1), how it feels about the brand on five levels (sentiment_level 0 to 4 and sentiment_score 0 to 1, null when the post is not about the brand), is it sarcastic, which aspect it talks about (taste_or_quality, price_or_value, availability_or_delivery, health_or_safety, advertising_or_campaign, customer_service, none), and did the author buy or use it (first_hand). intent: what the author is mainly doing (label: asking_for_recommendation, comparing_options, switching_away, complaining, promoting, news_or_discussion, other, or null when unsure, with confidence), whether they read as a potential buyer rather than a seller (buyer, seller), how pressing the need is (urgency, 0 to 3), and, when you pass offer=, whether your offer would plausibly help them (fits_offer). niche: which of the 33 niches of the published taxonomy sc-niche-v1 the caption belongs to, or personal_no_niche, or other (label, confidence, taxonomy), with label null when the caption is too thin to tell or the pick is unsure. It reads the caption only, not the video. quality: how much checkable detail the caption carries (fact_density 0 to 3), whether it mainly asks for likes, replies, shares, follows or tags (engagement_bait, 0 to 1), whether it is written to provoke anger as a way to get engagement (rage_bait, 0 to 1; about the writing, never the side taken), whether it only repeats someone else's news or view (secondhand, 0 to 1), and what the post is mainly doing (post_aim: inform, opinion, sell, entertain, provoke, other). exclude=engagement_bait drops posts whose engagement_bait is 0.8 or higher. ai_style: writing patterns common in AI-generated text, judged on the text alone. It does not say who wrote the post, and a person who writes in this style is flagged too. Six patterns are judged, each a probability (0 to 1): binary_contrast ("This is not X. It's Y."), faux_insight ("Here's what nobody tells you" and throat-clearing openers), colon_reveal ("The best part: it learns."), profound_kicker (ends on an aphorism or mic-drop line), puffery ("marks a pivotal moment", "highlighting the importance of") and rhetorical_setup ("What if I told you", a question the post answers itself, "Not a X. Not a Y. A Z."). Four are counted in the text: slop_words (words and phrases such as delve, leverage, robust, game changer, at the end of the day), em_dashes, emoji_bullets (lines that start with an emoji or a symbol bullet) and one_line_paragraphs (paragraphs that are one short line). pattern_count is how many patterns are present (a judged pattern at 0.8 or higher, slop_words 1 or more, em_dashes 2 or more, emoji_bullets 2 or more, one_line_paragraphs 5 or more) and markers lists their names, comma-separated. A post with less than about eight words of text is only counted, with every judged pattern 0. The judged patterns read the first 400 and the last 150 characters of a long post, and the counts read the whole post. exclude=ai_style drops posts with 2 or more patterns. ai_written: how likely the post was written by an AI model, from SocialCrawl's own text classifier, reading the opening of the post (about 1,000 characters): p (0 to 1, calibrated), level (likely (p 0.95 or higher), possible (p 0.71 or higher) or unlikely) and model (the model name; a new name means a new scale). A post under 8 words is too short to read: p null, level too_short. It is a probability, not proof: a person can be flagged and an AI-written post can pass, and the shorter the post the less certain it is. exclude=ai_written drops likely posts. injection flags text that addresses an AI system and tries to direct it (flagged, p); it never drops or rewrites a row. A post that could not be judged carries labels: null. Posts already labelled for anyone are free, and so is a cached page. data.labels reports what was judged and billed.
- exclude (optional, string): Optional CSV (engagement_bait, ai_style, ai_written). engagement_bait, only with label=quality, drops posts whose engagement_bait probability is 0.8 or higher. ai_style, only with label=ai_style, drops posts that show 2 or more of its writing patterns (pattern_count 2 or higher). ai_written, only with label=ai_written, drops posts whose level is likely (p 0.95 or higher), or whose p reaches your ai_written_threshold; a post too short to read is never dropped. Dropped posts are listed in data.labels.dropped_ids. A post that could not be judged is never dropped.
- ai_written_threshold (optional, string): Optional, only with label=ai_written and exclude=ai_written. The p (0 to 1) at which exclude=ai_written drops a post. Default 0.95 (level likely). 0.71 also drops possible posts: more AI posts go, and more posts by people with them.
- brand (optional, string): Required by label=mention, ignored otherwise. The brand or product the posts are judged against, up to 300 characters, for example brand=Buldak. Without it label=mention is skipped with the warning label_mention_needs_brand and is not billed.
- brand_description (optional, string): Optional, with label=mention. One plain sentence saying what the brand is, for example brand_description=Samyang's spicy instant noodle brand. Use it when the name is also an ordinary word or another company's name.
- offer (optional, string): Optional, with label=intent. One or two plain sentences saying what you sell, up to 300 characters, for example offer=A web design agency that builds websites for small businesses. Adds fits_offer to every post; without it fits_offer is null.
- label_evidence (optional, enum: 1): Optional, only with label=. When 1, every labelled row also carries computed.labels_evidence.<preset> = { quote, sentence_index }: the sentence in the row that most clearly shows the label, copied verbatim. Absent or null when no single sentence shows it.
- since (optional, string): Only rows published on or after this date: YYYY-MM-DD (midnight UTC) or an ISO 8601 timestamp. Rows older than it are left off each page. This list is not in date order, so the walk does not stop at the date: keep following `next_cursor`, and each page costs what a page costs even when every row on it is filtered out. A row with no date is kept. `published_after`, `start_date`, `from_date`, `publish_time` (`yesterday`, `this-week`, `this-month`, `last-3-months`, `last-6-months`) and `max_age_days` are accepted and applied as `since`.
- until (optional, string): Only rows published on or before this date: YYYY-MM-DD (that whole UTC day is included) or an ISO 8601 timestamp. Newer rows are left off each page and the walk does not stop, so older pages follow. A row with no date is kept. `published_before`, `end_date` and `to_date` are accepted and applied as `until`.
- max_pages (optional, integer): 1 to 50 (default 1, or 50 when `max_items` or `max_credits` is sent). Walk up to this many pages in one call and return the rows from all of them. Each page walked is billed exactly as one call to this endpoint (a cached page is free) and counts against your rate limit; its charge and any refund are recorded under the page's own request id, listed in `data.walk.request_ids`. The walk stops early at the last page, or when the request's time budget runs low; `data.walk.stopped` says why (`end`, `max_pages`, `max_items`, `max_credits`, `time_budget`, `page_error`) and `pagination.next_cursor` continues from where it stopped.
- max_items (optional, integer): 1 to 5000. Get more in one call: keep walking pages until the response holds at least this many rows, then stop with `data.walk.stopped: "max_items"`. The last page is returned whole (it was billed), so `data.items` can hold up to one page more than this, and `pagination.next_cursor` resumes right after it without skipping a row. Walks up to 50 pages unless `max_pages` says fewer; each page is billed as one call (a cached page is free). Add `max_credits` to cap the spend. A long walk can stop on the time budget first (`"time_budget"`): send the returned cursor with the same params to continue.
- max_credits (optional, integer): 1 to 100000. The most this call may spend, across every page it walks. A page starts only while the credits already spent plus that page's price (or, when higher, the most any page of this walk has cost) still fit, so the call never goes past it; the walk then stops with `data.walk.stopped: "max_credits"` and the cursor to resume. Sent alone, it walks as far as the budget goes (up to 50 pages). A value below the price of one page is a free 400 whose `error.details.page_credits` names that price. Credits a row join (`include=`) adds are known only once a page has run, so page 1 is checked against its own price.
Worth sending:
- label=sponsored: Marks each post that is paid or sponsored, at no extra credits; intent and niche are free too.
- relevance=filter: Drops rows that are not about your query, at no extra credits; relevance=score keeps them and adds a score.
curl -G "https://www.socialcrawl.dev/v1/naver/cafearticle/search" \
  --data-urlencode "query=주식" \
  --data-urlencode "label=sponsored" \
  --data-urlencode "relevance=filter" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/kin/search
Search Naver KnowledgeiN (지식iN)
Credit cost: 1 (standard)
Parameters:
- query (required): Free-text search term (UTF-8). Required.
- display (optional, integer): Number of items to return per page. Defaults to 10, cap 100; `display=101` returns 400.
- start (optional, integer): 1-indexed offset for pagination. Defaults to 1, cap 1000; `start=1001` is a 400 that costs nothing, and the page that reaches rank 1000 is the last one (`has_more: false`). Past rank 100, `sim` and `point` pages repeat Naver's top 100: those rows are removed and counted in `_warnings` (`recycled_rows`), and an emptied page is free. Page deeper with `sort=date`.
- sort (optional, enum: sim | date | point): Sort order. Accepted values: sim|date|point. Defaults to `sim` (relevance) when the corpus supports sort. Values outside this corpus's domain are rejected with a 400 before any upstream call. `date` lists the newest first, but Naver serves only the first 1,000 rows of a query (`start` caps at 1000), so on a common word it reaches back hours or days, not months, and on a multi-word query Naver also loosens the match.
- group (optional, enum: question): Send `question` to get one row per question. Naver returns one row per answer, so a question with three matching answers fills three rows; grouped, its best-ranked row stands for it and carries `answers_on_page` and `post.ext.answer_nos`. Grouping is within the page, the price is unchanged, and paging still advances by the rows Naver served.
- max_pages (optional, integer): 1 to 50 (default 1, or 50 when `max_items` or `max_credits` is sent). Walk up to this many pages in one call and return the rows from all of them. Each page walked is billed exactly as one call to this endpoint (a cached page is free) and counts against your rate limit; its charge and any refund are recorded under the page's own request id, listed in `data.walk.request_ids`. The walk stops early at the last page, or when the request's time budget runs low; `data.walk.stopped` says why (`end`, `max_pages`, `max_items`, `max_credits`, `time_budget`, `page_error`) and `pagination.next_cursor` continues from where it stopped.
- max_items (optional, integer): 1 to 5000. Get more in one call: keep walking pages until the response holds at least this many rows, then stop with `data.walk.stopped: "max_items"`. The last page is returned whole (it was billed), so `data.items` can hold up to one page more than this, and `pagination.next_cursor` resumes right after it without skipping a row. Walks up to 50 pages unless `max_pages` says fewer; each page is billed as one call (a cached page is free). Add `max_credits` to cap the spend. A long walk can stop on the time budget first (`"time_budget"`): send the returned cursor with the same params to continue.
- max_credits (optional, integer): 1 to 100000. The most this call may spend, across every page it walks. A page starts only while the credits already spent plus that page's price (or, when higher, the most any page of this walk has cost) still fit, so the call never goes past it; the walk then stops with `data.walk.stopped: "max_credits"` and the cursor to resume. Sent alone, it walks as far as the budget goes (up to 50 pages). A value below the price of one page is a free 400 whose `error.details.page_credits` names that price. Credits a row join (`include=`) adds are known only once a page has run, so page 1 is checked against its own price.
curl -G "https://www.socialcrawl.dev/v1/naver/kin/search" \
  --data-urlencode "query=코로나 증상" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/local/search
Search Naver Local (장소 검색)
Credit cost: 1 (standard)
Parameters:
- query (required): Free-text search term (UTF-8). Required.
- display (optional, integer): Number of places to return. This corpus caps at 5 and **defaults to 1**, so omitting it returns a single place. Values above 5 are silently clamped by Naver rather than rejected.
- sort (optional, enum: random | comment): Sort order. Accepted values: random|comment (display max 5, start max 1). Defaults to `sim` (relevance) when the corpus supports sort. Values outside this corpus's domain are rejected with a 400 before any upstream call.
curl -G "https://www.socialcrawl.dev/v1/naver/local/search" \
  --data-urlencode "query=강남역 카페" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/image/search
Search Naver Image
Credit cost: 1 (standard)
Parameters:
- query (required): Free-text search term (UTF-8). Required.
- display (optional, integer): Number of items to return per page. Defaults to 10, cap 100; `display=101` returns 400.
- start (optional, integer): 1-indexed offset for pagination. Defaults to 1, cap 1000; `start=1001` is a 400 that costs nothing, and the page that reaches rank 1000 is the last one (`has_more: false`). Past rank 100, `sim` and `point` pages repeat Naver's top 100: those rows are removed and counted in `_warnings` (`recycled_rows`), and an emptied page is free. Page deeper with `sort=date`.
- sort (optional, enum: sim | date): Sort order. Accepted values: sim|date. Defaults to `sim` (relevance) when the corpus supports sort. Values outside this corpus's domain are rejected with a 400 before any upstream call. `date` lists the newest first, but Naver serves only the first 1,000 rows of a query (`start` caps at 1000), so on a common word it reaches back hours or days, not months, and on a multi-word query Naver also loosens the match.
- filter (optional, enum: all | large | medium | small): Restrict to an image size band: `all` (default), `large`, `medium`, or `small`.
- since (optional, string): Only rows published on or after this date: YYYY-MM-DD (midnight UTC) or an ISO 8601 timestamp. Rows older than it are left off each page. This list is not in date order, so the walk does not stop at the date: keep following `next_cursor`, and each page costs what a page costs even when every row on it is filtered out. A row with no date is kept. `published_after`, `start_date`, `from_date`, `publish_time` (`yesterday`, `this-week`, `this-month`, `last-3-months`, `last-6-months`) and `max_age_days` are accepted and applied as `since`.
- until (optional, string): Only rows published on or before this date: YYYY-MM-DD (that whole UTC day is included) or an ISO 8601 timestamp. Newer rows are left off each page and the walk does not stop, so older pages follow. A row with no date is kept. `published_before`, `end_date` and `to_date` are accepted and applied as `until`.
- max_pages (optional, integer): 1 to 50 (default 1, or 50 when `max_items` or `max_credits` is sent). Walk up to this many pages in one call and return the rows from all of them. Each page walked is billed exactly as one call to this endpoint (a cached page is free) and counts against your rate limit; its charge and any refund are recorded under the page's own request id, listed in `data.walk.request_ids`. The walk stops early at the last page, or when the request's time budget runs low; `data.walk.stopped` says why (`end`, `max_pages`, `max_items`, `max_credits`, `time_budget`, `page_error`) and `pagination.next_cursor` continues from where it stopped.
- max_items (optional, integer): 1 to 5000. Get more in one call: keep walking pages until the response holds at least this many rows, then stop with `data.walk.stopped: "max_items"`. The last page is returned whole (it was billed), so `data.items` can hold up to one page more than this, and `pagination.next_cursor` resumes right after it without skipping a row. Walks up to 50 pages unless `max_pages` says fewer; each page is billed as one call (a cached page is free). Add `max_credits` to cap the spend. A long walk can stop on the time budget first (`"time_budget"`): send the returned cursor with the same params to continue.
- max_credits (optional, integer): 1 to 100000. The most this call may spend, across every page it walks. A page starts only while the credits already spent plus that page's price (or, when higher, the most any page of this walk has cost) still fit, so the call never goes past it; the walk then stops with `data.walk.stopped: "max_credits"` and the cursor to resume. Sent alone, it walks as far as the budget goes (up to 50 pages). A value below the price of one page is a free 400 whose `error.details.page_credits` names that price. Credits a row join (`include=`) adds are known only once a page has run, so page 1 is checked against its own price.
curl -G "https://www.socialcrawl.dev/v1/naver/image/search" \
  --data-urlencode "query=한라산" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/webkr/search
Search Naver Web (웹문서)
Credit cost: 1 (standard)
Parameters:
- query (required): Free-text search term (UTF-8). Required.
- display (optional, integer): Number of items to return per page. Defaults to 10, cap 100; `display=101` returns 400.
- start (optional, integer): 1-indexed offset for pagination. Defaults to 1, cap 1000; `start=1001` is a 400 that costs nothing, and the page that reaches rank 1000 is the last one (`has_more: false`). Past rank 100, `sim` and `point` pages repeat Naver's top 100: those rows are removed and counted in `_warnings` (`recycled_rows`), and an emptied page is free. This corpus ignores `sort`, so `sort=date` does not avoid the repeats.
- sort (optional, string): Sort order. Accepted values: (sort ignored). Defaults to `sim` (relevance) when the corpus supports sort. Values outside this corpus's domain are rejected with a 400 before any upstream call.
- max_pages (optional, integer): 1 to 50 (default 1, or 50 when `max_items` or `max_credits` is sent). Walk up to this many pages in one call and return the rows from all of them. Each page walked is billed exactly as one call to this endpoint (a cached page is free) and counts against your rate limit; its charge and any refund are recorded under the page's own request id, listed in `data.walk.request_ids`. The walk stops early at the last page, or when the request's time budget runs low; `data.walk.stopped` says why (`end`, `max_pages`, `max_items`, `max_credits`, `time_budget`, `page_error`) and `pagination.next_cursor` continues from where it stopped.
- max_items (optional, integer): 1 to 5000. Get more in one call: keep walking pages until the response holds at least this many rows, then stop with `data.walk.stopped: "max_items"`. The last page is returned whole (it was billed), so `data.items` can hold up to one page more than this, and `pagination.next_cursor` resumes right after it without skipping a row. Walks up to 50 pages unless `max_pages` says fewer; each page is billed as one call (a cached page is free). Add `max_credits` to cap the spend. A long walk can stop on the time budget first (`"time_budget"`): send the returned cursor with the same params to continue.
- max_credits (optional, integer): 1 to 100000. The most this call may spend, across every page it walks. A page starts only while the credits already spent plus that page's price (or, when higher, the most any page of this walk has cost) still fit, so the call never goes past it; the walk then stops with `data.walk.stopped: "max_credits"` and the cursor to resume. Sent alone, it walks as far as the budget goes (up to 50 pages). A value below the price of one page is a free 400 whose `error.details.page_credits` names that price. Credits a row join (`include=`) adds are known only once a page has run, so page 1 is checked against its own price.
curl -G "https://www.socialcrawl.dev/v1/naver/webkr/search" \
  --data-urlencode "query=기후변화" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/errata
Correct a mistyped Korean search query (오타변환)
Credit cost: 1 (standard)
Parameters:
- query (required): The possibly-mistyped search term (UTF-8). Required.
curl -G "https://www.socialcrawl.dev/v1/naver/errata" \
  --data-urlencode "query=네이볘" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/adult
Check whether a Korean search term is adult-only (성인 검색어 판별)
Credit cost: 1 (standard)
Parameters:
- query (required): The search term to classify (UTF-8). Required.
curl -G "https://www.socialcrawl.dev/v1/naver/adult" \
  --data-urlencode "query=성인영화" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/search-trend
Get Naver search-volume trend for Korean keywords (검색어트렌드)
Credit cost: 5 (advanced)
Parameters:
- keywords (optional, string): Comma-separated Korean keywords, up to 20, combined into ONE trend series (volumes summed, not compared). Send this or `groups`.
- groups (optional, string): Up to 5 keyword groups, each its own series on one shared scale: separate groups with `|` and the keywords inside a group with `,`, for example `손 다한증,발 다한증|미라드라이|암내,액취증`. Each series is named by its first keyword. Up to 20 keywords per group. Send this or `keywords`, not both.
- start_date (optional, string): Window start, `YYYY-MM-DD`. Defaults to 12 months before `end_date`. Clamped up to 2016-01-01, the earliest data Naver holds.
- end_date (optional, string): Window end, `YYYY-MM-DD`. Defaults to today.
- time_unit (optional, enum: date | week | month): Aggregation bucket: `date`, `week`, or `month` (default `month`).
- group_name (optional, string): Label for the `keywords` series in the response. Defaults to the first keyword. Not used with `groups`.
- device (optional, enum: pc | mo): Restrict to `pc` or `mo` (mobile). Omit for both.
- gender (optional, enum: f | m): Restrict to `f` or `m`. Omit for both.
- ages (optional, string): Comma-separated age-band codes, `1` to `11`: `1` 0-12, `2` 13-18, `3` 19-24, `4` 25-29, `5` 30-34, `6` 35-39, `7` 40-44, `8` 45-49, `9` 50-54, `10` 55-59, `11` 60+. Several codes are SUMMED into one audience, not split. NOTE this endpoint uses different codes from `shopping-insight/*`, which takes 10/20/30/40/50/60.
- Constraint: one of keywords, groups (at least one required)
curl -G "https://www.socialcrawl.dev/v1/naver/search-trend" \
  --data-urlencode "keywords=삼성전자,갤럭시" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/shopping-insight/category
Get Naver Shopping click trend for a category (쇼핑인사이트)
Credit cost: 5 (advanced)
Parameters:
- category_code (required): Comma-separated Naver Shopping category ids, up to 3 (e.g. `50000000` for 패션의류). Required. When `breakdown` is set, only the first id is used. Naver publishes no category-list API, so see the platform guide for the id table.
- start_date (optional, string): Window start, `YYYY-MM-DD`. Defaults to 12 months before `end_date`. Clamped up to 2017-08-01, the earliest Shopping Insight data Naver holds.
- end_date (optional, string): Window end, `YYYY-MM-DD`. Defaults to today.
- time_unit (optional, enum: date | week | month): Aggregation bucket: `date`, `week`, or `month` (default `month`).
- device (optional, enum: pc | mo): Restrict to `pc` or `mo` (mobile). Omit for both.
- gender (optional, enum: f | m): Restrict to `f` or `m`. Omit for both.
- ages (optional, string): Comma-separated age buckets: `10`, `20`, `30`, `40`, `50`, `60`. NOTE these differ from `search-trend`, which takes 1-11.
- breakdown (optional, enum: device | gender | age): Split a single category by `device`, `gender`, or `age` instead of comparing several categories.
curl -G "https://www.socialcrawl.dev/v1/naver/shopping-insight/category" \
  --data-urlencode "category_code=50000000" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/shopping-insight/keyword
Get Naver Shopping click trend for keywords inside a category (쇼핑인사이트)
Credit cost: 5 (advanced)
Parameters:
- category_code (required): A single Naver Shopping category id to search within (e.g. `50000000`). Required.
- keyword (required): Comma-separated keywords, up to 5, each returned as its own comparable series inside that category. Required. When `breakdown` is set, only the first keyword is used.
- start_date (optional, string): Window start, `YYYY-MM-DD`. Defaults to 12 months before `end_date`. Clamped up to 2017-08-01, the earliest Shopping Insight data Naver holds.
- end_date (optional, string): Window end, `YYYY-MM-DD`. Defaults to today.
- time_unit (optional, enum: date | week | month): Aggregation bucket: `date`, `week`, or `month` (default `month`).
- device (optional, enum: pc | mo): Restrict to `pc` or `mo` (mobile). Omit for both.
- gender (optional, enum: f | m): Restrict to `f` or `m`. Omit for both.
- ages (optional, string): Comma-separated age buckets: `10`, `20`, `30`, `40`, `50`, `60`. NOTE these differ from `search-trend`, which takes 1-11.
- breakdown (optional, enum: device | gender | age): Split a single keyword by `device`, `gender`, or `age` instead of comparing several keywords.
curl -G "https://www.socialcrawl.dev/v1/naver/shopping-insight/keyword" \
  --data-urlencode "category_code=50000000" \
  --data-urlencode "keyword=코트" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/brief
One query across the Korean internet (5 Naver corpora) + optional digest.
Credit cost: 10 (override; tier advanced)
Parameters:
- query (required): Search query (Korean or any language).
- corpora (optional, string): CSV subset of news,blog,cafearticle,kin,webkr (default all five). `shop` was retired by Naver on 2026-07-31 and is rejected.
- display (optional, integer): Items per corpus (1-100, default 20).
- start (optional, integer): 1-indexed offset per corpus (1-1000, default 1). Prefer `cursor` for paging.
- sort (optional, string): sim (relevance, default) or date; kin also point.
- include (optional, string): Set to `digest` for an LLM English digest with translated quotes.
- cursor (optional, string): Opaque pagination token from a prior response's next_cursor.
curl -G "https://www.socialcrawl.dev/v1/naver/brief" \
  --data-urlencode "query=삼성전자" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/place/rank
Check where a place ranks on Naver Map for keywords (플레이스 순위)
Credit cost: 4-42 (metered): 4 credits per keyword, plus 2 credits when lat and lng are left out (the point is read from the place). The whole amount is refunded when the check fails.
Parameters:
- keywords (required): One to ten keywords, separated by commas, exactly as a person types them into Naver, e.g. `신사역 정형외과,강남 대상포진`.
- id (optional, string): The Naver Place id, e.g. 11669922 (the number in the place's link). Send this or url.
- url (optional, string): The place's link as Naver shows it: m.place.naver.com/hospital/11669922/home, map.naver.com/p/entry/place/11669922, or a naver.me share link. Send this or id.
- lat (optional, string): Latitude of the point to search from (send with lng). Leave both out to search from the place itself.
- lng (optional, string): Longitude of the point to search from (send with lat).
- top (optional, integer): How many of the top-ranked places to list with each result, 0 to 20 (default 5).
- ads (optional, boolean): Also check the paid 플레이스 광고 slots and report the place's `ad_rank` (default false).
- Constraint: one of id, url (at least one required)
curl -G "https://www.socialcrawl.dev/v1/naver/place/rank" \
  --data-urlencode "keywords=신사역 정형외과,강남 대상포진" \
  --data-urlencode "id=11669922" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/place/search
Search Naver Map places for a keyword, in Naver's order (플레이스 검색)
Credit cost: 4 (override; tier advanced)
Parameters:
- query (required): The keyword, exactly as a person types it into Naver Map, e.g. `신사역 정형외과`.
- lat (optional, string): Latitude of the point to search from (send with lng). Default: Gangnam Station.
- lng (optional, string): Longitude of the point to search from (send with lat).
- limit (optional, integer): How many organic places to return, 1 to 100 (default 100). The price is the same at any limit.
- ads (optional, boolean): Also return the paid 플레이스 광고 slots, after the organic places (default false).
curl -G "https://www.socialcrawl.dev/v1/naver/place/search" \
  --data-urlencode "query=신사역 정형외과" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/place/reviews
Get the newest visitor reviews of a Naver Place (방문자 리뷰)
Credit cost: 5-9 (metered): 5 credits per page. The default labels (sentiment, issue) are free. label=reports, incentivized or injection holds 4 extra credits and refunds down to 1 credit per started 25 reviews that were newly judged (the first 100 reviews of a page are judged): reviews already labelled are free, a page where nothing could be judged refunds the whole extra, and a cached page is free.
Parameters:
- id (optional, string): The Naver Place id, e.g. 11669922 (the number in the place's link). Send this or url.
- url (optional, string): The place's link as Naver shows it: m.place.naver.com/hospital/11669922/home, map.naver.com/p/entry/place/11669922, or a naver.me share link. Send this or id.
- sort (optional, enum: recent | recommended): `recent` (default): newest visits first. `recommended`: Naver's own order.
- label (optional, string): Optional CSV of SocialCrawl labels to add to every review. Without this param every page already carries sentiment and issue, free; label= adds the labels you name to them (the defaults keep running). sentiment and issue are free when asked for too; reports, incentivized and injection add 1 credit per started 25 newly judged reviews. judgments=off (or label=none) turns the default labels off. They read the review's title and text, never its stars. sentiment: how the reviewer feels, on five levels (level 0 to 4, score_0_1, confidence), plus rating_mismatch, true when the words clearly contradict the star rating (a glowing text under 1 star, a furious one under 5), null when the review has no rating. issue: the main problem the review reports (label: product_defect, sizing_or_fit, shipping_or_delivery, customer_service, price_or_value, missing_feature, other, or none, with confidence; label is null when unsure). reports (needs reports=): does this review say that the thing you describe happened (p, 0 to 1). incentivized: does the reviewer say they got the product free, discounted or rewarded for the review (p), and did the text carry a disclosure such as 'in exchange for my honest review', Vine or 체험단 (disclosed). injection flags text that addresses an AI system and tries to direct it (flagged, p); it never drops or rewrites a row. These are signals to read, never a verdict on a review or a reviewer. A review that could not be judged carries labels: null. Reviews already labelled for anyone are free, and so is a cached page. data.labels reports what was judged and billed.
- reports (optional, string): Required by label=reports, ignored otherwise. What to look for, in plain words, up to 300 characters, for example reports=the battery drains quickly or reports=배송이 늦음. Every review gains labels.reports.p, the probability that it says so, in any wording or language. Without it label=reports is skipped with the warning label_reports_needs_reports and is not billed.
- judgments (optional, enum: on | off): Optional, on (the default) or off. By default every row gains free SocialCrawl judgments (computed.labels, and computed.relevance on search endpoints), reported in data.labels (mode default) and data.relevance (origin default), each with a status (complete, partial or skipped) and pending: the rows still being judged when the page was sent, which carry null now and are filled on your next call or cached read. Default judgments never add credits, never change an existing field, and never drop or reorder a row. off returns the page exactly as before, with none of those keys. label=none does the same.
- dry_run (optional, enum: 1): Optional. When 1, return a cost preview for this labelled or relevance-filtered request without fetching the page or judging any row. data.estimate reports rows_expected, rows_cached, label_credits_min, label_credits_max and base_credits. 0 credits charged.
- label_evidence (optional, enum: 1): Optional, only with label=. When 1, every labelled row also carries computed.labels_evidence.<preset> = { quote, sentence_index }: the sentence in the row that most clearly shows the label, copied verbatim. Absent or null when no single sentence shows it.
- fit (optional, enum: goal): Optional. When goal, keep the rows and fields needed for the goal you pass in goal= (plus any that are uncertain, and the first and last), and replace the rest with a stub. data.held_back lists the held ids and a recall id that re-reads the full page from cache at no extra charge. Without this param the page is unchanged.
- goal (optional, string): Required by fit=goal. What you are trying to do, in your own words, up to 300 characters.
- fit_tokens (optional, integer): Optional, only with fit=goal. Soft cap on how much of the page to keep, in tokens. Uncertain blocks and the first and last block are kept even if they exceed it.
- since (optional, string): Only rows published on or after this date: YYYY-MM-DD (midnight UTC) or an ISO 8601 timestamp. Rows older than it are left off the page, and the page that reaches one ends the walk: `next_cursor` is not returned and `pagination.stopped_at` is `since`. Sending it sets the newest-first sort shown in this endpoint's sort parameter (any other sort value is a free 400). The page still costs what a page costs, so you pay for the pages you walk and no more. A row with no date is kept. `published_after`, `start_date`, `from_date`, `publish_time` (`yesterday`, `this-week`, `this-month`, `last-3-months`, `last-6-months`) and `max_age_days` are accepted and applied as `since`.
- until (optional, string): Only rows published on or before this date: YYYY-MM-DD (that whole UTC day is included) or an ISO 8601 timestamp. Newer rows are left off each page and the walk does not stop, so older pages follow. A row with no date is kept. `published_before`, `end_date` and `to_date` are accepted and applied as `until`.
- Constraint: one of id, url (at least one required)
Worth sending:
- label=issue: Names the product issue each review raises, at no extra credits; sentiment is free too.
curl -G "https://www.socialcrawl.dev/v1/naver/place/reviews" \
  --data-urlencode "id=11669922" \
  --data-urlencode "label=issue" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/place
Get a Naver Place card (플레이스 정보)
Credit cost: 3 (override; tier advanced)
Parameters:
- id (optional, string): The Naver Place id, e.g. 11669922 (the number in the place's link). Send this or url.
- url (optional, string): The place's link as Naver shows it: m.place.naver.com/hospital/11669922/home, map.naver.com/p/entry/place/11669922, or a naver.me share link. Send this or id.
- Constraint: one of id, url (at least one required)
curl -G "https://www.socialcrawl.dev/v1/naver/place" \
  --data-urlencode "id=11669922" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/blog
Get a Naver blog's card and visitor counter (블로그 정보)
Credit cost: 1 (standard)
Parameters:
- handle (optional, string): The Naver blog id, e.g. flock8979 (the part after blog.naver.com/). Send this or url.
- url (optional, string): The blog's link, e.g. blog.naver.com/flock8979 or m.blog.naver.com/flock8979. Send this or handle.
- Constraint: one of handle, url (at least one required)
curl -G "https://www.socialcrawl.dev/v1/naver/blog" \
  --data-urlencode "handle=blogpeople" \
  -H "x-api-key: YOUR_API_KEY"

## GET /v1/naver/blog/posts
List every post of a Naver blog, newest first (블로그 글 목록)
Credit cost: 1 (standard)
Parameters:
- handle (optional, string): The Naver blog id, e.g. flock8979 (the part after blog.naver.com/). Send this or url.
- url (optional, string): The blog's link, e.g. blog.naver.com/flock8979 or m.blog.naver.com/flock8979. Send this or handle.
- page (optional, integer): Page number, 30 posts a page (default 1). Prefer `cursor`.
- since (optional, string): Only rows published on or after this date: YYYY-MM-DD (midnight UTC) or an ISO 8601 timestamp. Rows older than it are left off each page. This list is not in date order, so the walk does not stop at the date: keep following `next_cursor`, and each page costs what a page costs even when every row on it is filtered out. A row with no date is kept. `published_after`, `start_date`, `from_date`, `publish_time` (`yesterday`, `this-week`, `this-month`, `last-3-months`, `last-6-months`) and `max_age_days` are accepted and applied as `since`.
- until (optional, string): Only rows published on or before this date: YYYY-MM-DD (that whole UTC day is included) or an ISO 8601 timestamp. Newer rows are left off each page and the walk does not stop, so older pages follow. A row with no date is kept. `published_before`, `end_date` and `to_date` are accepted and applied as `until`.
- max_pages (optional, integer): 1 to 50 (default 1, or 50 when `max_items` or `max_credits` is sent). Walk up to this many pages in one call and return the rows from all of them. Each page walked is billed exactly as one call to this endpoint (a cached page is free) and counts against your rate limit; its charge and any refund are recorded under the page's own request id, listed in `data.walk.request_ids`. The walk stops early at the last page, or when the request's time budget runs low; `data.walk.stopped` says why (`end`, `max_pages`, `max_items`, `max_credits`, `time_budget`, `page_error`) and `pagination.next_cursor` continues from where it stopped.
- max_items (optional, integer): 1 to 5000. Get more in one call: keep walking pages until the response holds at least this many rows, then stop with `data.walk.stopped: "max_items"`. The last page is returned whole (it was billed), so `data.items` can hold up to one page more than this, and `pagination.next_cursor` resumes right after it without skipping a row. Walks up to 50 pages unless `max_pages` says fewer; each page is billed as one call (a cached page is free). Add `max_credits` to cap the spend. A long walk can stop on the time budget first (`"time_budget"`): send the returned cursor with the same params to continue.
- max_credits (optional, integer): 1 to 100000. The most this call may spend, across every page it walks. A page starts only while the credits already spent plus that page's price (or, when higher, the most any page of this walk has cost) still fit, so the call never goes past it; the walk then stops with `data.walk.stopped: "max_credits"` and the cursor to resume. Sent alone, it walks as far as the budget goes (up to 50 pages). A value below the price of one page is a free 400 whose `error.details.page_credits` names that price. Credits a row join (`include=`) adds are known only once a page has run, so page 1 is checked against its own price.
- Constraint: one of handle, url (at least one required)
curl -G "https://www.socialcrawl.dev/v1/naver/blog/posts" \
  --data-urlencode "handle=flock8979" \
  -H "x-api-key: YOUR_API_KEY"