Table of Contents

Every page gets a contents rail on the right, built from its headings. You are looking at one now — on a wide enough screen.

What appears

Headings from h2 to h4:

# Page Title ← not listed; it is the page, not a section within it

## A section ← listed

### A subsection ← listed, indented

#### A detail ← listed, indented further

##### Too deep ← not listed

h1 is skipped because it names the page as a whole, and the rail already sits on that page. h5 and below are skipped to keep the rail readable.

Pages with fewer than two headings get no rail at all — a contents list with one entry is noise.

Anchors

Every heading receives an id derived from its text, so each section is directly linkable:

## Getting Started → #getting-started

## What's new in v2 → #whats-new-in-v2

Link to one from anywhere:

[Jump to setup](#setup)
[From another page](/getting-started/quick-start#prerequisites)
[[quick-start#prerequisites]]

Search results link to these anchors too, which is what lets a hit open at the right section.

Scroll tracking

The entry for whichever section you are reading is highlighted as you scroll. The active section is chosen by position — the last heading to pass the top of the viewport — so a long section stays highlighted after its heading scrolls out of view.

Where it appears

The rail shows on screens 1280px and wider. Below that the article takes the full width; the headings are still in the page and still linkable, just not listed alongside.

Generated at build time

The heading list is extracted while your Markdown is compiled, so it ships in the HTML rather than being assembled by a script after the page loads. Only the scroll highlighting runs in the browser.

Next

• Search — find pages by their contents
• Wiki Links — link between pages by name