Перейти к содержимому
TakoAPI
Назад к навыкам

html2md

от saikatkumardey

Web & Frontend Development
—Ещё нет оценок
Ваша оценка

Кратко

Convert HTML pages to clean, agent-friendly markdown using Readability + Turndown.

Описание

Что делает этот навык

Converts HTML pages to clean markdown using Mozilla Readability for content extraction and Turndown for conversion. Strips navigation, ads, footers, cookie banners, and other noise. Supports URL fetching, local files, stdin, token budgeting, and JSON output.

Readability-based extraction removes nav and ad noise that raw HTML fetchers leave in, so token usage stays low and content arrives clean for LLM consumption.

Когда его использовать

  • Extracting article content from news sites for LLM summarization
  • Converting documentation pages to markdown for offline reference
  • Feeding cleaned web content to sub-agents in research pipelines
  • Token-budgeted page extraction to fit LLM context windows
  • Batch scraping blog posts in cron-based agent workflows

Пример сценария

ВВОД

User asks: Summarize the key points from https://paulgraham.com/greatwork.html, keeping it under 2000 tokens

АГЕНТ
  1. 1

    Run html2md --max-tokens 2000 https://paulgraham.com/greatwork.html

  2. 2

    Readability isolates the main article body, discarding nav, footers, and sidebars

  3. 3

    Turndown converts the cleaned HTML to markdown

  4. 4

    Post-processing removes remaining noise and truncates at the 2000-token budget

  5. 5

    Agent receives clean markdown and generates the summary

ВЫВОД

Clean markdown of the article body truncated to 2000 tokens, ready for summarization

Установка

clawhub install saikatkumardey/html2md

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

Ссылки