Naver Real Estate Search
Analysis
This skill appears aligned with searching and monitoring Naver real-estate listings, but users should notice its local scraper dependency, external Naver requests, and saved watch/cache files.
Findings (3)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.
이 스킬은 로컬 upstream clone을 래핑한다. - `tmp/naverland-scrapper`
The skill discloses reliance on a local upstream scraper clone that is not part of the manifest or install metadata, so its exact version/provenance is outside the packaged artifacts.
Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.
SEARCH_URL = "https://search.naver.com/search.naver?...&query={query}" ... html = sre._request_text(url)The code sends search strings to Naver and retrieves web content, which is expected for a Naver real-estate search skill but means search terms leave the local environment.
로컬 JSON 파일에 watch rule 저장 ... `last_seen` / `events` / dedupe ... 저장 파일: `data/watch-rules.json`, `data/candidate-cache.json`, `references/candidate-seeds.json`
The skill intentionally persists watch criteria, observed listings, and candidate aliases in local JSON files that can influence future alerts and search results.
