Question 1

What does FusedFrames mean by reasoning capture?

Accepted Answer

It is the record of why your team did something, not just what they did. The desktop app watches the work as it happens. When a decision looks different from the way similar work was handled before, it asks a single question. That way the explanation is captured in the moment rather than lost.

Question 2

When does the app actually ask a question?

Accepted Answer

Only when a step cannot be explained by what it has seen before. The first few days teach it your team's routine, so expect more questions early on and very few once the app knows what normal looks like. Predictable work is always recorded silently.

Question 3

Do I have to answer straight away?

Accepted Answer

No. You can dismiss the question and keep working, though dismissing it skips it for good. When you do answer, you can type or hold the talk button and speak, whichever is faster in the moment.

Question 4

What does the app record while it is watching?

Accepted Answer

It records clicks, keystrokes, copy and paste, drag and drop, selected text, app and window names and small captures of parts of the screen. It does not record passwords, card numbers, API keys or webcam video, and the microphone is only used while you hold the talk button to answer a question. Sensitive content is stripped on your device before anything leaves it.

Question 5

Can I stop recording for sensitive or personal work?

Accepted Answer

Yes. Right-click the FusedFrames icon in your menu bar or system tray and choose Stop recording, then Start recording when you are back. On Team and Enterprise, recording rules can also start and stop recording for you based on the application, the window title or the time of day. On Enterprise, an admin can mark a rule as required, and recording stays on while that rule matches. While recording is off, nothing is captured.

Question 6

Where do my answers end up?

Accepted Answer

Each answer is attached to the recording it explains and written into the documents built from it, right next to the steps. Teammates read the reasoning in the dashboard and AI agents get it over MCP, the CLI and the API, so the why never gets separated from the what.

Question 7

How is this different from maintaining a runbook or agents.md?

Accepted Answer

A static file contains what someone remembered to write on the day they wrote it, and it goes stale from that moment on. Reasoning capture works the other way round. The work is recorded as it happens, the app asks about the parts a document would never have predicted, and repeat recordings keep everything current without anyone maintaining a file.

Question 8

Can other people in my workspace see my answers?

Accepted Answer

Only if you share them. Answers travel with the recording they explain. Recordings and the libraries they feed stay private to their owner until they are shared with the workspace or with selected members. Anything stripped as sensitive on your device, or done while recording was off, was never uploaded at all.

Press record once. Let the documents write themselves.

Clear steps, backed by the real recording

Captures the reasoning behind every step

Frequently asked questions

Start recording today