CaptaHow to turn a screen recording into a step-by-step guide on a Mac
Five steps from a recording to a document a colleague can follow, with each click boxed and numbered.
A step-by-step guide is a screenshot per click, a sentence per screenshot, and a mark on the button you pressed so the reader can find it. Writing one by hand takes an afternoon. Capta makes it from the same recording it makes the video from: the clicks, the typing and the scrolling are already on the recording's event track, so the guide is planned before you open it. You read it, fix a sentence or two, and export. Everything runs on your Mac, including the model that names buttons the app left unnamed.
Record the task once, at a normal pace
Press ⌥⌘R and do the task the way you would show it to someone at your desk. Click the controls you mean and pause before the important ones. Stop with ⌥⌘R. A desktop app, a remote desktop session and a browser tab all work the same way, because the guide comes from what happened on screen rather than from a browser extension.
Typed text stays out of the guide unless Show typed text is on in Settings › Recording. A password field is blacked out on its still either way.
Switch the editor to Guide
In the editor, click Guide in the Video | Guide switch at the top, or press ⇧⌘G. The steps are already there: one per click with a box on the element and a numbered badge, one per typing run, one per shortcut, one per long scroll, and a last still of the result. The sentence under each comes from the element's name: "Click Save", "Type into the Client field", "Press ⌘S". When an app gives no name, Capta reads the label off the picture.
Fix what a reader would trip on
Click a sentence to rewrite it, or type a tip under it. Switch a still between Before and After the click, nudge it half a second either way, or press Frame to zoom and drag the picture by hand. Hide a step that is noise, Merge two that belong together, and add a note for the one thing the recording could not show. Redact lets you drag a rectangle over anything private on a single still; masks drawn in the Edit tab cover every still. ⌘Z undoes any of it.
Pick a template, then adjust
The sidebar's Template tab has ten starting points, from a knowledge-base look with focused stills to a client handout with whole windows and a title page. Stills holds the crop, the box or spotlight on the element, the badge shape, the line weight and the colour. Document holds the type, including a Google Fonts family for the web page, the text size, the page theme and the title page's subtitle, author, date, footer and cover picture. Words hands every unedited sentence to the writer chosen in Settings › Models, and can add a line under each step about what it does.
Export, copy, or push to Hudu
PDF for a client or an email. Markdown with a folder of PNGs for a repository, Notion or Obsidian. HTML as one self-contained page with a header, a sidebar of the steps and a search box. Copy puts rich text with the pictures on the clipboard for Google Docs, Confluence or Mail. With a Hudu address and API key in Settings › Integrations, Push to Hudu creates the knowledge base article with the stills attached, and a second push updates the same article.
Every format is free: PDF, Markdown, HTML, the clipboard and the Hudu push, with no watermark on the stills.
What makes the stills readable
- The frame just before the click. The control is in its resting state, hover and all, so the reader sees what to aim at. Switch to After when the result is the point.
- The box comes from the accessibility tree. Capta records the element under every click while you record, so the box lands on the button rather than near it.
- The crop keeps the words around the element. A focused still grows to keep a line it would have cut, the label beside a field, or the dialog the button sits in.
- Labels read off the picture. Remote desktops and canvases give no names. The on-device text recognizer reads the word under the pointer and the sentence uses it.
- Retina stills at half size. A still pastes at its on-screen size and prints sharp. A fifty-step PDF stays under ten megabytes.
Common questions
Does it need a browser extension? No. Capta records the screen, so a remote desktop window, a terminal, a native app and a browser tab all come out as the same kind of guide. New recordings also switch on the accessibility tree inside Chrome and Electron apps, so web buttons get their names.
Can I get the video and the guide from one take? Yes. Record once. Export writes the video and Export guide writes the guide from the same recording, and both share the masks you drew.
Where is the guide stored? As guide.json inside the recording's bundle in ~/Movies/Capta, next to the video and the edit. Your sentences, hidden steps and redactions stay with the recording.
What about IT Glue or Confluence? Paste. Copy puts the steps and pictures on the clipboard as rich text, which those editors take. Hudu has a direct push.
One shortcut to record, one click to export.