Feedback your coding agent can act on.
Your users click what is wrong. Loopback takes the screenshot, lets them draw on it, files the issue in your tracker, and keeps the conversation in your app until it is fixed.
One loop, from the click to the fix.
- The user clicks the broken partLoopback takes a screenshot, and they draw on it to show what is wrong.
- An issue opens in your trackerThe screenshot, the page and the browser details go in with it.
- Your agent picks it upWhatever coding agent you already use can work the issue. So can a developer.
- The user gets the reply in your appQuestions and answers go back and forth in the widget, right where they reported it.
It keeps going until the problem is fixed.
Everything around the request.
- You see the screenshot with the user's drawing on it.
- You know the page, the element, the browser, who reported it and when.
- The whole history is in one timeline, and every entry is copied to the tracker.
- Need more? Ask from the item, and the user answers in the widget.
Built for your coding agent.
- It picks the issue up from the tracker it already works in.
- It can ask the user a question through Loopback and get the answer back the same way.
- When the fix ships, it marks the item ready for the user to check.
Next up: the interview.
- Loopback's own agent will ask the follow-up questions right after the report, while the user still has the page open.
- In development.
Your tracker, your agent, a thin widget.
- Your tracker is the record.
- Issues, comments and the closing pull request stay in GitHub. There is nothing to migrate onto.
- Bring the agent you already use.
- The loop runs on plain issue comments and one small API.
- A widget, not a suite.
- One button on your app and one inbox for the team. No portal, no help desk.
Put a feedback button on your app that your agent can answer.
One script tag in your app, one repository to file into.