> ## Documentation Index
> Fetch the complete documentation index at: https://livingcontext.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# Editing steps

> Refine step text, fix highlights, annotate screenshots, and frame the perfect shot before publishing.

A raw capture is usually close, but rarely perfect. Before publishing, you can polish every step — in the side panel, or in the full-page view when you want more room to work.

## Working with steps

* **Edit descriptions inline** — click any step's text and rewrite it in place.
* **Reorder** — drag steps into the order that reads best.
* **Delete** — remove steps that don't belong in the guide.
* **Bulk select** — select multiple steps to delete them in one action.

<Tip>
  You don't have to write great descriptions by hand — publishing runs AI over the capture and writes a terse description for every step. Focus your editing on removing noise and getting the visuals right.
</Tip>

## Fixing the click position

Each click step shows a highlight on the screenshot marking where the action happened. If the highlight isn't quite where readers should look, open the click-position editor and drag the spotlight to the right spot. The same position is used when the guide is replayed with [Guide Me](/capture/guide-me).

## Annotating screenshots

The annotation editor draws directly on the screenshot. Available tools:

| Tool             | Use it for                                           |
| ---------------- | ---------------------------------------------------- |
| Box              | Framing a region of the page                         |
| Circle / ellipse | Calling out a button or field                        |
| Arrow            | Pointing from a label to a target                    |
| Freehand pen     | Anything the shapes can't express                    |
| Text label       | Short callouts on the image itself                   |
| Numbered marker  | Ordered points of interest within one screenshot     |
| Redaction fill   | Solid cover over sensitive content                   |
| Blur region      | Softly obscuring content that should stay unreadable |

Annotations render identically everywhere — in the editor, in the published doc, and in downloads — so what you see while editing is exactly what readers get.

<Warning>
  Redaction fills and blur regions are permanently baked into the published images. Once you publish, the covered pixels cannot be recovered from the doc. Automatic protections also apply before you ever reach this editor — see [Privacy and redaction](/capture/privacy-and-redaction).
</Warning>

## Crop and zoom

Use crop and zoom to frame each screenshot on what matters — a single form section instead of the whole page, for instance.

Framing is non-destructive:

* Readers see the cropped, zoomed result in the published guide.
* The source screenshot is kept intact, so you can re-open the editor later and reframe without re-capturing.

## When you're happy

Once the steps read well and the screenshots are framed and clean, move on to [publishing](/capture/publishing) — one click turns the capture into a change request in your docs repo.

<Note>
  Editing doesn't end at publish. In the change request, every step image has a full action menu — annotate, blur, redact, crop, **replace the image**, set alt text, show a typing step as a GIF or a clip from the session recording, and adjust the click target — so you can keep refining right up to merge.
</Note>