article-extractor_skill

Overview

This skill extracts the main readable content from article, blog, and tutorial URLs and saves it as clean text files. It removes navigation, ads, newsletter prompts, and other clutter so you get only the article content and headings. It names and cleans the file using the article title and shows a preview after saving.

How this skill works

It checks for extraction tools in priority order (reader → trafilatura → fallback curl+parser). The skill downloads the page, runs the best available extractor, and post-processes the text (whitespace cleanup, optional comment/table removal). It then builds a safe filename from the article title, saves the .txt file, and prints a preview and location confirmation.

When to use it

• You have a URL and want the article saved as plain text.
• You need article content without ads, navigation, or signups.
• You want an offline readable copy of a blog post or tutorial.
• You need a quick export for note-taking, summarization, or archiving.
• You want the extractor to pick the best available tool automatically.

Best practices

• Prefer reader (Mozilla Readability) for most news and blogs.
• Use trafilatura for complex sites, academic pieces, or non-English content.
• Always inspect the preview (first 10–15 lines) before finalizing.
• Clean filenames to remove special characters and limit length (<100 chars).
• If extraction fails, try the alternate tool or use the fallback curl method.

Example use cases

• Extract and save a news article for offline reading with reader.
• Download a long-form tutorial and remove ads for research notes.
• Batch-extract multiple blog posts using trafilatura for reliable structure.
• Fallback-extract simple pages when no extraction tools are available.
• Quickly produce plain-text input for summarization or NLP tasks.