Перейти к содержимому
TakoAPI
Назад к навыкам

zoomin-scraper-recklessop

от recklessop

Web & Frontend Development
476Число загрузок по данным clawskills.sh
0Просмотры
0 лайковВойдите, чтобы поставить лайк навыку
—Ещё нет оценок
Ваша оценка

Кратко

Scrape documentation content from Zoomin Software portals using Playwright browser automation to handle dynamic.

Описание

Что делает этот навык

Scrapes documentation content from Zoomin Software portals using a headless Chromium browser. Standard HTTP fetching fails on these portals because content loads dynamically via JavaScript. The skill navigates to each URL, waits for the main article element to render, then extracts the text.

Executes JavaScript and waits for DOM rendering, capturing content that standard HTTP scrapers miss entirely on Zoomin portals.

Когда его использовать

  • Batch downloading Zoomin-hosted vendor documentation
  • Archiving product docs before portal access expires
  • Building a local knowledge base from dynamic documentation sites
  • Extracting text from docs that return empty content via requests/BeautifulSoup
  • Processing a list of documentation URLs into plain text files

Пример сценария

ВВОД

User asks: 'Scrape these 40 Zoomin documentation URLs and save the content to a local folder'

АГЕНТ
  1. 1

    Creates a text file with one target URL per line

  2. 2

    Runs run_scraper.sh with the URL file path, output directory, and virtual environment path

  3. 3

    Playwright launches headless Chromium and navigates to each URL in sequence

  4. 4

    Waits for the article element with id 'zDocsContent' to finish loading

  5. 5

    Saves the extracted text to individual files in the output directory

ВЫВОД

Plain text files in the output directory, each containing the rendered documentation body from one URL

Требования

  • Playwright Python package (pip install playwright)
  • Chromium browser binaries (playwright install chromium)

Установка

clawhub install recklessop/zoomin-scraper-recklessop

Выполните команду CLI или вставьте промпт в Claude Code, Cursor, Codex — любой кодинг-агент, которым пользуетесь.

Ссылки