How reports work

What the person reporting a bug actually sees, click by click.

1. They click the launcher

The corner button opens a small menu with two choices: 📷 Screenshot or 🎥 Record video. A screenshot is right for “this looks wrong”; a video is right for “this breaks when I do X, then Y”.

2. The browser asks to share their screen

Both options use the browser's built-in screen sharing, so the reporter sees the browser's own permission prompt and picks what to share — the tab, a window, or the whole screen. Nothing is captured until they allow it. A screenshot grabs a single frame and stops immediately; a video records until they end it.

Screen sharing isn't supported by most mobile browsers, so the widget is effectively desktop-only for now. Reporters on phones won't be able to capture.

3. They describe it and hit send

A compose panel shows what they captured with a note box — “What went wrong?” — and Send / Discard buttons. The note is optional (up to 2,000 characters) but it usually carries the intent: what they expected versus what happened.

4. It lands with you

The report appears under Recordings, grouped by project.
Videos are stored as .webm, screenshots as .jpg; uploads are capped at 100 MB.
On the Pro plan, the report arrives with the technical context attached and is posted to the project's Slack channel if one is connected.

Audio

Video reports can include tab or microphone audio, but only if you enable Capture audio in that project's Widget settings and the reporter allows it in the share prompt. It is off by default.