Turn your clicks into a step-by-step guide

No. Recording, editing, redaction and export all happen on your machine, and guides are stored as plain files in your user folder. Wombat talks to one address: its public releases page, to check for a new version. If there is one it downloads in the background, and asks before installing anything.

This website is a separate thing from the app, and it does keep one number: the clicks people make here — pressing record, and anything you click while it's running — add up to a shared total. That's the only thing sent, there's no identifier attached to it, and it says nothing about anything you record with the app itself. Even writing to us doesn't touch this site — those buttons open your own mail client, so the message goes from you to us and past nothing in between.

Anything on screen. Wombat captures the display and hooks input system-wide, so desktop software, spreadsheets and internal clients record exactly like web apps do. Label quality depends on how well an app exposes accessibility information — native Windows and Chromium-based apps are generally good; some canvas-drawn or Java/Qt interfaces expose almost nothing and fall back to generic text.

Password fields are never read: the accessibility helper refuses them at the source, and the step text is redacted again independently, so a value can't leak through either path. For everything else there's manual redaction — drag over a region to pixelate it, and “apply permanently” to rewrite the file so the original pixels are gone. Exports always bake redactions in.

Yes — no account, no trial, no paid tier. It's a desktop app that runs locally, so there's no server bill to pass on.

Because it isn't code-signed. SmartScreen flags any binary from a publisher it doesn't recognise, regardless of what's in it. Choose More info → Run anyway. You can also verify what you downloaded against the checksums on the release page.

Recording, screenshots, the editor, redaction and export all work. Smart step labels don't — they're built on Windows UI Automation, so on macOS every step reads “Click here” until the equivalent native helper is written. macOS also needs Screen Recording and Accessibility permissions granted in System Settings, and the disk image is unsigned, so the first launch needs right-click → Open.

All of it. Rewrite any instruction, reorder or delete steps, insert a step you forgot, drag the click marker to the right spot, drop a mis-detected outline, and toggle the zoom per step or across the whole guide.

A single self-contained HTML file, with every image inlined — so it renders from a mailbox, a shared drive or a wiki attachment with nothing else alongside it. That's deliberately the only format, rather than a half-decent version of four.