Evidence capture for bug reports
Collect annotated screenshots, console errors, network activity and browser context with website bug reports, so your team can investigate from one ticket.
Included on every plan, screen recording with it
Annotated screenshot, drawn by the reporter on their own screen
Console errors and network calls, with secrets scrubbed
How it works
- 01
The reporter marks what is wrong
The widget captures the page they are looking at and hands them a pen to circle it. No screenshot tool, no file to attach.
- 02
The widget attaches the context
Console logs, network activity, page URL, viewport and browser travel with the report, because nobody remembers to include them.
- 03
Engineering opens one ticket
The ticket brings together the available screenshot, replay, logs and browser context. Capture settings, browser support and performance budgets decide what is available; the report can still be sent without evidence.
The first hour stops going to questions
The console and the network calls from the seconds before the report are on the ticket, grouped and levelled: the uncaught TypeError, the two POSTs that came back 500, the clicks in between. Reproduction starts on the first read instead of after three replies.
You can name everything it keeps
The context a ticket carries is a closed list: page path, first seen, browser, OS, device, screen, viewport, time zone and language. Request and response bodies are never captured, query strings are cut from every URL before it is stored, and IP addresses are kept salted-hashed.
Exact behaviour
- Capture
- Annotated screenshot, console, network, URL, viewport, browser
- Retention
- Captured logs 30 days; idle sessions 90 days; tickets are never removed by retention
- Privacy
- authorization · cookie · password · token · secret scrubbed from logs; bodies never captured; IPs salted-hashed
- Browsers
- Chromium, WebKit, Firefox; desktop and mobile
- Plan
- Every plan, screen recording included
- MCP tool
- get_ticket · search_logs · get_errors
Works with
Questions teams ask before installing
Does the screenshot include what the user typed?
The screenshot is the page as the reporter saw it, and they draw on it and send it themselves, so anything visible on their screen can be in it. Session replay is the part that is masked: it never records typed characters.
Are request and response bodies captured?
No. Network activity is recorded as the call and its result, never the payload, and authorization, cookie, password, token and secret are stripped from captured logs.
What does capture cost the host page?
Shadow DOM contains the widget’s styles. Separate CPU, memory, network and rendering budgets limit its work, and it reduces or suspends capture when the host page is under pressure.
The thinking behind it
Get the next bug report with its evidence attached.
Free includes evidence capture, with unlimited seats and 5 projects.