How does the Markdown TOC generator work?

Paste your markdown document. The parser walks the source line-by-line, skipping fenced code blocks (``` and ~~~), and extracts H1–H6 ATX headings (# Title) plus setext-style headings (text followed by === for H1 or --- for H2). It then generates a clickable nested list of anchor links using the slug algorithm for your chosen platform. You can set a min/max depth, skip the first heading, switch between bullet and ordered lists, and insert the result back into your document at a [TOC] or marker.

Why do anchor slugs differ between GitHub, GitLab, pandoc, and Bitbucket?

Each platform has its own heading-to-anchor slug rules. GitHub and GitLab lowercase the heading, strip punctuation except hyphens, and join words with hyphens. Pandoc strips all punctuation including existing hyphens. Bitbucket preserves underscores. This tool ships per-platform slugify implementations so every generated link resolves correctly when rendered on the matching platform — switch the dropdown and the slugs update live.

How are duplicate headings handled?

If your document has two headings with the same text (e.g. two ## Examples sections), the tool appends a -1 suffix to the second occurrence, -2 to the third, and so on — matching GitHub's behavior. So the first link points to #examples, the second to #examples-1, the third to #examples-2. Each generated anchor matches what the platform's renderer will emit.

Will # inside a code fence be mistaken for a heading?

No. The parser tracks fenced code blocks (both ``` and ~~~) and skips lines inside them, so a comment like `# this is a comment` inside a code block is not picked up as a heading. Setext-style headings (text on one line, === on the next) are also detected, and indented headings inside the document body are handled correctly.

What extra features does this tool have versus other TOC generators?

(1) Per-platform slug algorithms (GitHub, GitLab, pandoc, Bitbucket, generic). (2) Duplicate-heading disambiguation with -1, -2 suffixes. (3) Code-fence-aware parsing (skips ``` and ~~~ blocks). (4) Setext heading support (=== and ---). (5) Configurable min/max depth (H1–H6). (6) Skip-first-heading option. (7) Bullet or ordered list output. (8) Insert at [TOC] marker. (9) Insert at / marker. (10) Update existing TOC in place. (11) Optional anchors for renderers without auto-IDs. (12) Emoji-aware heading handling. (13) Stats (heading count per level). (14) History (localStorage, last 20). (15) Shareable URL. 100% client-side — your documents never leave the browser.

Developer Runs in your browser

Markdown TOC Generator

Generate a clickable Markdown table of contents from any document's headings. Per-platform slug algorithm (GitHub, GitLab, pandoc, Bitbucket, generic), duplicate-heading disambiguation (-1, -2…), code-fence-aware parsing, setext heading support, depth range control, skip-first-heading, ordered or bullet list style, and insert/update at a [TOC] or  marker. 100% client-side.

100% Private Works Offline Instant

Platform

Min level

Max level

List style

Skip first headingEmit <a name> anchors

Markdown source

Output

Headings

Duplicates

Platform

GitHub

TOC chars

Paste a markdown document to generate a TOC

Headings (H1–H6) become a clickable nested list with anchor links. Code fences and setext headings are handled.

Privacy: All parsing and TOC generation runs locally. Your documents never leave the browser. History is stored in localStorage on this device only.

How to use

Enter your input in the tool above.
Adjust any options to your preference.
Use the Copy or Download buttons to save the result.
Everything happens locally — your data never leaves your browser.

All tools