naver-land-search

Naver Land Apartment Search

Search for an apartment on Naver Real Estate (fin.land.naver.com) and collect structured information.

Usage

/naver-land-search 대방대림
/naver-land-search 잠실엘스

Arguments

• apartment name (required): Korean apartment name to search for

Constraints

• MUST use --headed mode (headless is blocked by bot detection on this site)
• MUST navigate via URL with query parameter (keyboard input search may not trigger results)
• All playwright-cli commands run via the Bash tool

Workflow

Step 1: Open Browser

Open a headed Chrome browser and navigate directly to the search URL.

playwright-cli open --headed "https://fin.land.naver.com/search?q=<URL-encoded apartment name>"

Step 2: Verify Search Results

Take a snapshot and read it to check results.

playwright-cli snapshot

• If 0 results: report to user that no results were found.
• If 1 result: proceed to Step 3 with that result.
• If multiple results: present the list to the user using AskUserQuestion and let them choose. Each result shows the type (아파트/빌라), name, and address.

Step 3: Select Apartment

Click the chosen search result button. The page navigates to the apartment detail page.

Step 4: Collect Basic Info

From the detail page snapshot, extract:

Field	Location in snapshot
세대수	listitem near heading (e.g. "1,628세대")
준공년도/년차	listitem with date and 년차 (e.g. "1993. 11. (33년차)")
용적률	listitem (e.g. "용적률 272%")

Step 5: Select Area Type

The default area type may not match what user wants. Check the area toggle button and:

1. Click the area toggle to open the dropdown
2. Find the area type closest to the user's target (default: 전용 59m²)
3. Click the matching option
4. Wait for page data to refresh

If the user specifies a target area, use that instead of the default 59m².

Step 6: Collect Transaction Data

From the refreshed snapshot, extract:

Field	Location in snapshot
매매/전세	The currently selected trade type toggle
선택된 면적	The area toggle label (e.g. "84.47㎡ (59.84)")
최근 실거래 금액	strong element after "최근 실거래"
최근 실거래 날짜	generic element with date after "최근 실거래"
3년 내 최고	text after "3년 내 최고" (price + contract date)
3년 내 최저	text after "3년 내 최저" (price + contract date)
출처 URL	The apartment detail page URL (e.g. https://fin.land.naver.com/complexes/364) — use the base path without query params

Step 7: Close Browser

playwright-cli close

Step 8: Write to Google Spreadsheet

Use the gws-sheets skill to write collected data to a spreadsheet.

Spreadsheet ID: 1DmclS3MGoM1uRNxG0iUo5KX86sYn7KNjP2lmVgpRYRs

1. Read the spreadsheet to check if a header row exists:

gws sheets +read --spreadsheet=1DmclS3MGoM1uRNxG0iUo5KX86sYn7KNjP2lmVgpRYRs --range="A1:K1"

2. If the header row is missing, append it first:

gws sheets +append --spreadsheet=1DmclS3MGoM1uRNxG0iUo5KX86sYn7KNjP2lmVgpRYRs --values '아파트명,주소,세대수,준공,용적률,면적,최근실거래금액,최근실거래일,3년내최고,3년내최저,출처URL'

3. Append the data row:

gws sheets +append --spreadsheet=1DmclS3MGoM1uRNxG0iUo5KX86sYn7KNjP2lmVgpRYRs --json-values '[["<아파트명>","<주소>","<세대수>","<준공>","<용적률>","<면적>","<최근실거래금액>","<최근실거래일>","<3년내최고>","<3년내최저>","<출처URL>"]]'

Column mapping:

Column	Value example
아파트명	대방대림
주소	서울시 동작구 대방동
세대수	1628세대
준공	1993.11 (33년차)
용적률	272%
면적	전용 59.84㎡ (공급 84.47㎡)
최근실거래금액	14억 5000만원
최근실거래일	2026.2.13
3년내최고	14억 5000만원 (계약 26년 2월)
3년내최저	8억 4500만원 (계약 23년 6월)
출처URL	https://fin.land.naver.com/complexes/364

Step 9: Report

Present the collected data in a structured table:

## <아파트명> (<주소>)

### 단지 기본 정보
| 항목 | 값 |
|------|-----|
| 세대수 | ... |
| 준공 | ... (..년차) |
| 용적률 | ...% |

### 매매 실거래 (전용 ..m²)
| 항목 | 값 |
|------|-----|
| 최근 실거래 | ...원 (YYYY.MM.DD) |
| 3년 내 최고 | ...원 (계약 YY년 M월) |
| 3년 내 최저 | ...원 (계약 YY년 M월) |

Spreadsheet updated: https://docs.google.com/spreadsheets/d/1DmclS3MGoM1uRNxG0iUo5KX86sYn7KNjP2lmVgpRYRs

Notes

• This site blocks headless browsers. Always use --headed mode.
• URL-based search (?q=...) is more reliable than typing into the search box.
• If the browser is already open, close it first before opening a new headed session.
• The skill defaults to 매매 (sale) trade type and 전용 59m² area. If the user specifies different preferences, adjust accordingly.
• Google Sheets writing requires gws CLI with authenticated Google account.