Free HTML to Text Cleaner

Web scraping pipelines

1. For one-off scrapes, paste the page's body HTML into this tool and copy the output.
2. For repeated scrapes, use the html-to-text npm package or BeautifulSoup .get_text() in Python.
3. Always strip whitespace and deduplicate empty lines after the cleaner step.

WordPress to Webflow migration

1. Export WordPress posts (Tools > Export > Posts).
2. For each post, copy the post_content from the XML export.
3. Run through this cleaner to get plain text. Paste into Webflow CMS rich-text fields, re-add formatting via Webflow's editor.

Notion as content source

1. If you copy from Notion, the clipboard often includes HTML formatting alongside the visible text.
2. Paste into this cleaner to strip the HTML, leaving just the text.
3. Useful when piping Notion content into other systems that choke on Notion's rich-text format.

AI / LLM input prep

1. Article HTML often has 30-60% of bytes in tags and scripts.
2. Clean to text before sending to OpenAI, Anthropic, or your own LLM. You save tokens and reduce input noise.
3. For long articles, also chunk the cleaned text to fit context windows (typically 100K-200K tokens depending on model).

Email HTML extraction

1. Save email HTML (most clients let you "Show Original" or "View Source" on a message).
2. Paste into this cleaner. The output is the email's actual content without table-layout artifacts.
3. Strip newsletter boilerplate (header logo, footer unsubscribe) by hand — it appears as text but is not part of the article.

Output looks like a wall of text with no breaks

Cause: Your input HTML used <br> tags or <span> for line breaks instead of block-level <p> tags.

Fix: Manually add line breaks after sentences, or run the input through an HTML formatter first to introduce proper structure, then re-clean.

Script or style content appears in the output

Cause: You used a different tool that does not strip <script> / <style> contents.

Fix: This tool removes them. If you see code in your output, you used a different tool. Re-run with the Grigora HTML to Text Cleaner.

HTML entities (& or <) appear in output

Cause: Some entities remained because the source had non-standard encoding.

Fix: Run the output through a separate HTML decoder, or paste into a browser console: document.querySelector("textarea").value = decoded; The cleaner handles common entities; obscure ones may slip through.

Text is duplicated

Cause: Your input had visually-hidden duplicate content (e.g., a desktop nav and a mobile nav with identical text).

Fix: After cleaning, deduplicate by hand or with a text-deduplication tool. The cleaner sees raw HTML; CSS-hidden duplicates are not filtered.

Lists do not look like lists

Cause: List items lose their bullets in plain text by definition.

Fix: After cleaning, manually add bullet characters (- or &middot;) at the start of each line if you need visual list formatting.

Image alt text is missing

Cause: The cleaner strips <img> tags entirely by default, including alt attributes.

Fix: For alt-text preservation, use an HTML-to-Markdown converter (which writes alt text as ). Or strip just images manually before cleaning.

Tables come through as scrambled text

Cause: HTML tables with inconsistent widths or merged cells flatten poorly to plain text.

Fix: For tabular data, convert to CSV or Markdown table syntax first. Plain-text cleaning of tables produces readable but ugly output.

Tool fails on very large input (10MB+)

Cause: Browser memory limits.

Fix: Split into chunks of 1-2MB. Or use html-to-text Node CLI for files over 10MB — it processes in streams and handles arbitrary size.