.html → .md converter
PUB 2026-08-01UPD 2026-10-27VERIFY Standards and official sources APP $ html2md ✓ Converts in your browser without uploading the file to a server.Convert copied web content or HTML to Markdown (.md). Paste source code, paste content copied from a browser, or open an .html file. View the result in a new window, save it as a file, or copy it directly.
text/html) that preserves headings, lists, tables, and links. If you paste without formatting, the text is kept as-is.
You can paste repeatedly to add more content. New content goes at the cursor position. Use Enter to create a blank line and paste there. Otherwise, content is placed after the block containing the cursor. To start over, use the button at the top labeled Clear.
| Category | HTML | Markdown | Type |
|---|
Using this tool
When it helps
Use this tool to turn an HTML guide into an editable Markdown draft, or to organize headings, lists, and links copied from a web page as working notes. If you have the source, put it in INPUT — HTML code. Use Paste web content for material copied from a browser. Open an HTML file from your device can also read a saved document.
The two input methods have separate drafts and results. You can supply the same content as source code and as copied web content to compare what remains. With Append when opening a file enabled, the HTML code tab adds content at the end; Paste web content inserts it after the block containing the cursor. If the cursor is on a blank line, content goes there; if it is outside the input area, content goes at the end. With the option off, the file replaces the current tab's content. Reading and conversion take place in your browser; the tool does not upload the file to a server to perform the conversion.
Understanding results
RESULT — From HTML code and From pasted web content show separate outputs. View in a new window, Save as an MD file, and Copy apply to the tab currently displayed. Inspect the character, line, heading, and link counts below the result. Compare it with the source to check that heading levels, list order, and link labels remain as intended.
Search RULES by tag or explanation to see what becomes Markdown and what is omitted. Common identifies basic syntax; GFM identifies extensions controlled by the option; Removed means retaining content alone or excluding an item. Notices above RESULT list some omitted items. The absence of a notice does not establish that every piece of information from the original has been preserved.
Measured case
The measured HTML contained the heading 점검, bold text 굵게, list items 첫째 and 둘째, a link labeled 홈, and a table with headers 항목 and 값 and a row containing A and 1. With GFM on and reference links off, the Markdown result contained 87 characters across 12 lines, with 1 heading and 1 link.
Turning only GFM off changed the table into two text rows containing 항목 and 값, then A and 1, and displayed a notice that the table formatting was omitted. The other structures remained. The result contained 66 characters across 12 lines, with 1 heading and 1 link.
Questions and answers
Why did formatting disappear when I pasted web content?
Paste web content reads the HTML formatting supplied by the browser. Copying or pasting without formatting supplies only text, so the tool cannot recover the original headings or table structure. If source HTML is available, compare the HTML code input with the pasted version.
What happens to tables and checkboxes when GFM is off?
Tables are exported as text containing their rows, while task list checkmarks and strikethrough formatting are omitted. Table text and list content remain, but the same presentation does not. Choose the option after checking whether the destination editor supports GFM.
What does collecting links as references do?
It places reference numbers at links in the body and gathers destination URLs at the end of the result. This can help when editing a draft with long links. Copy the definitions at the end as well, so the references still resolve.
Will conversion reproduce the original web page?
The converter carries over structures such as headings, paragraphs, and lists, rather than reproducing page layout or CSS design. It excludes scripts, styles, and comments. It also does not identify the Markdown that originally generated the HTML or restore its exact choice of symbols.
Further details
How it operates
The browser HTML parser reads the input into a tree of document elements. The converter removes excluded elements and attributes, then turns blocks and their inline emphasis, links, and code into Markdown in sequence. It escapes characters where necessary so they are not mistaken for formatting, organizes paragraph spacing, and adds reference definitions if that option is enabled.
Images retain their addresses and alternative text in Markdown image syntax; video and audio addresses become links. Font loading information is omitted. This does not download media files or embed them in the result. Relative addresses can point to a different location after you move the output, so check them where the resulting document will actually be used.
Standards and references
CommonMark Spec 0.31.2 describes basic output syntax, the GitHub Flavored Markdown Spec describes extensions, and the HTML Living Standard describes HTML parsing. They do not prescribe one unique reverse conversion to the original Markdown. Consult the on screen conversion rules for the elements this tool keeps or excludes.
Constraints
A file may contain up to 300,000 UTF-16 code units, and appending checks the combined total. Files that exceed the limit are rejected instead of truncated; the limit is not enforced for typing or pasting. Tables with merged cells or complex nesting may lose their original arrangement. Compare cell relationships and content order after conversion.
The tool removes elements and attributes that load external resources before inserting content into the page. However, the browser may request external addresses while parsing HTML, so complete blocking is not guaranteed. Opening the result elsewhere can also request retained addresses. No file upload does not mean that every network request is absent.
Standards: CommonMark Spec 0.31.2 (Markdown syntax) · GitHub Flavored Markdown Spec (tables, task lists, strikethrough) · HTML Living Standard (HTML parsing)