Ditto: Step-by-Step Guides & Tutorials de Skyforge AI

Click record, do the thing, get a polished guide.

Ditto watches what you do in the browser and turns it into a step-by-step guide with
annotated screenshots — no manual cropping, no writing each step by hand. Every click,
keystroke and page navigation becomes a step, with the clicked element highlighted and
zoomed in automatically.

Everything runs inside your browser. There is no backend, no account and no tracking.

WHAT IT DOES

• Auto-capture — clicks, form input, keyboard shortcuts, clipboard actions, drag events
  and navigations. Rapid clicks on nearby elements are merged so guides stay readable,
  and clicks are captured before the page navigates away, so nothing is lost on single-
  page apps or full page loads.
• Smart Blur — emails, phone numbers, SSNs, credit cards, IP and MAC addresses are
  detected and masked before the screenshot is taken. Anything the patterns miss, you can
  select by hand and mask across every screenshot it appears in.
• Guide Me — replay any guide live on the real page. Ditto highlights the next element,
  tracks your progress and advances as you interact. Useful for onboarding someone
  without sitting next to them.
• Voice narration (optional) — talk through the workflow while you record and Ditto turns
  what you said into the step descriptions, matched to the action each sentence belongs to.
• AI descriptions (optional) — turn "Click login-button" into "Click the Login button to
  sign in". Generated from a short text snippet of page context, never from your
  screenshots.
• Guide editor — crop, annotate and redact screenshots, rewrite text inline, add headings
  and notes between steps, reorder or bulk-delete, and roll back through version history.
• Export anywhere — PDF, DOCX, HTML, Markdown, MP4/WebM video or GIF. All generated in
  the browser; nothing is uploaded to produce them.
• Six languages — English, Spanish, French, German, Portuguese (Brazil) and Chinese
  (Simplified).

PRIVACY

Your guides, screenshots and settings stay in your browser's own storage. Ditto has no
servers, so there is nothing for your data to be sent to.

Two optional features talk to a third party, and only if you turn them on: AI descriptions
and voice transcription send text or audio to a provider you choose, using your own API
key. You can also point Ditto at a model server you run yourself, in which case nothing
leaves your machine at all. Site icons are fetched from Google's favicon service, which
sends that site's domain and nothing else.

No analytics, no telemetry, no account, no ads, and your data is never used to train
anything.

OPEN SOURCE

MIT licensed. Read the code, file an issue, or build it yourself:
https://github.com/tryskyforge/ditto