> For the complete documentation index, see [llms.txt](https://docs.spara.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.spara.com/solutions-engineer/in-the-meeting/live-session-screen.md).

# Live session screen

Every part of Samantha's live session screen: answer cards, transcript, participants, voice controls, screenshare controls and the pop-out.

The live session screen is where you follow Samantha during a call. It shows answer cards for the prospect's questions as they come up, the transcript and who's in the call. For a Samantha who speaks, it's also where you turn her voice on and off and control what she shares on screen.

Spara opens this screen as soon as you send Samantha from [**Dispatch**](https://app.spara.co/meeting-agents). To come back to it, during or after the call, click **Open** next to her under **Recent meetings**. See [Send Samantha to a meeting](/solutions-engineer/getting-started/send-samantha-to-a-meeting.md).

{% hint style="info" %}
The screenshots on this page use the local Acme test account and a made-up call, with an agent named Spara. A Samantha on "Hey Samantha" looks the same, with her name in place of Spara's.
{% endhint %}

## The screen at a glance

This is a Samantha who speaks, sent with an AI Prospect and a product demo, a few minutes into a call.

<figure><img src="https://821105055-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FTMt4EjieWT1FfljMBJHI%2Fuploads%2Fgit-blob-219a43c16f4e899dab9d32f08577f8dd942ada98%2FLive%20Session%20Screen.png?alt=media" alt=""><figcaption><p>The live session screen during a call</p></figcaption></figure>

**Header**

1. **Status**: What Samantha is doing. See [#status](#status "mention").
2. **Meeting link**: Opens the meeting in a new tab.
3. **Pop out**: During the call, opens the cards in a small window that stays over your meeting. See [#pop-out](#pop-out "mention").
4. **Leave meeting**: Until the call ends, takes Samantha out of it after you confirm. Her cards and transcript stay here. An AI Prospect sent with her stays in the call; remove it with **Leave** under **Recent meetings**.

**Control bar**

5. **AI Prospect status**: Shows when you sent an AI Prospect with Samantha: its name and status, and the reason if it fails.
6. **Voice**: Whether Samantha is about to speak, speaking, or waiting, with **Turn on** or **Turn off**. Only for a Samantha who speaks. See [#voice](#voice "mention").
7. **Auto-show**: Lets Samantha put visuals on her screenshare by herself. See [#auto-show](#auto-show "mention").
8. **On screen**: What her screenshare shows right now, who put it there, and **Clear visual**. See [#on-screen](#on-screen "mention").

**Answer cards**

9. **Answer card**: The question it answers, the answer, and supporting points. The top card is outlined and in larger type. See [#answer-cards](#answer-cards "mention").
10. **Suggested visuals**: Visuals from your product demos that fit the answer. **Show** puts one on the screenshare.
11. **AI demo topic**: A product demo topic Samantha can demo. **AI Demo** starts a narrated demo of it. See [#ai-demos](#ai-demos "mention").

**Side panel**

12. **Participants**: Who's in the call and which side they're on, with **This is me** to fix who the rep is. See [#participants](#participants "mention").
13. **Transcript**: The call so far, with notes when the screenshare changes. See [#transcript](#transcript "mention").

The control bar only shows what applies. A Samantha who only listens has no voice or screenshare controls, and her cards have no visuals or demos. Lines like **Connecting…** or **Reconnecting…** also show there while the screen reaches the call.

## Status

The tag next to Samantha's name in the header.

| Status                     | Meaning                                                                   |
| -------------------------- | ------------------------------------------------------------------------- |
| **Joining**                | She is on her way into the meeting.                                       |
| **Waiting to be admitted** | She is in the waiting room. The host needs to admit her.                  |
| **Listening**              | She is in the call and listening.                                         |
| **Thinking**               | She is deciding whether what the prospect side just said needs a card.    |
| **Ended**                  | She left, or the meeting ended.                                           |
| **Failed**                 | She couldn't join, or her call failed. The reason shows under the header. |

Before the first card, the main area says what's happening: **Joining the meeting**, **Listening** ("Answers appear here when the prospect asks something"), that she couldn't join, or that the meeting ended with no answers.

## Answer cards

Samantha writes a card when someone on the prospect side raises something you need to address. The prospect side is everyone not marked **Rep**, an AI Prospect included.

* **What a card shows**: The question it answers in small grey text, the answer in bold, and supporting points under it.
* **Order**: Cards from the latest exchange come first, the most important on top. The top card is outlined and in larger type.
* **Cards grow**: When the prospect adds detail, Samantha updates the card in place rather than adding another.
* **Nothing is removed**: The cards stay as a running log of the call, and are still here after it ends.

For a Samantha who speaks, the cards are also what she answers from when she speaks.

## Voice

Only for a Samantha who speaks, while she's in the call.

| Voice                                 | Meaning                                                                                                   |
| ------------------------------------- | --------------------------------------------------------------------------------------------------------- |
| **Off · waiting for its wake phrase** | She won't speak until someone says her wake phrase or you click **Turn on**.                              |
| **On**                                | She answers what's meant for her: a question someone asks her by name, or a follow-up to her last answer. |
| **About to speak**                    | She has decided to answer, or the next step of an AI demo is coming.                                      |
| **Speaking**                          | She's talking in the meeting.                                                                             |

* **Turn off** stops her right away, including anything she's saying or about to say.
* **Turn on** wakes her without the wake phrase.
* When she was sent without **Sales Engineer leads the call**, she also turns herself off when someone dismisses her ("Thanks, Samantha") or after a quiet spell.
* When she leads the call, she starts on and only **Turn off** on this screen turns her off.

## Presenting

A Samantha who speaks presents on her own screenshare in the meeting. What she can show comes from the product demos you picked when you sent her.

### Auto-show

On, Samantha can put up the visual that fits what she's answering, by herself, just before she answers. Off, she only shows what someone asks for. Even with Auto-show on, she holds back while an AI demo runs, while someone else in the meeting is sharing, and for a short while after you change the screen yourself.

### On screen

The visual on the screenshare now, with its thumbnail and name, and **by Spara** (Samantha put it up) or **by you**.

* **Clear visual** takes it down and stops the screenshare. It also ends a running AI demo.
* A visual stays up until you clear it or someone asks Samantha to take it down. A visual Samantha put up also comes down when someone dismisses her while her voice is on.
* When the screenshare is up with nothing on it, **Stop sharing** ends it.
* **Starting screenshare…** or **Screenshare failed** shows next to it when that happens.

### Suggested visuals

A card can suggest up to two visuals. **Show** puts one on the screenshare, starting it if needed, and the tile reads **On screen** while it's up.

### AI demos

A card can offer a product demo topic. **AI Demo** has Samantha narrate a demo of that topic step by step, putting each step's visual on screen. An AI demo starts only when it's asked for: from **AI Demo** here, or when someone in the meeting asks Samantha for one.

Under the topic, the stack of its visuals ("Click to show your own") opens them all as **Show** tiles, so you can walk through them yourself instead.

<figure><img src="https://821105055-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FTMt4EjieWT1FfljMBJHI%2Fuploads%2Fgit-blob-c92727e278565ea0f09b688d8e57b125c1a76ba1%2FLive%20Session%20AI%20Demo.png?alt=media" alt=""><figcaption><p>An AI demo running</p></figcaption></figure>

1. **Voice**: **Speaking** while she narrates.
2. **AI demo**: The topic and its progress, for example "Step 1 of 2", with **Stop**. It reads **Starting…** while the demo gets ready.
3. **On screen**: The current step's visual, put up by Samantha.
4. **The card's demo**: Highlighted while its demo runs, with **Stop demo**.

Someone talking over the narration stops the demo, and so does clearing the visual.

## Participants

Everyone in the call, each tagged **Rep** or **Prospect**. **Host** marks the meeting host, and **Left** marks someone who has dropped off.

Spara marks you as the rep when your email or name matches someone in the call, and puts everyone else on the prospect side. If it picked the wrong person, click **This is me** on your row. That relabels your earlier transcript lines too, and from then on Samantha treats what you say as the rep's. **This is me** shows only while the screen is connected to the call.

## Transcript

The call so far, oldest first. Each line shows who spoke and their side: **Rep**, **Prospect**, or **AI** for Samantha herself. A line of hers that someone talked over ends with "(cut off)".

Notes between the lines record what the screenshare did, such as "Spara showed products" (Samantha put it up), "You cleared the visual" or "AI demo of Unified Vendor started". The transcript keeps up with new lines unless you scroll up to read.

## Pop-out

**Pop out** puts what you need at a glance into a small window you can keep over the meeting. Click **Close pop-out** in the header to close it.

<figure><img src="https://821105055-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FTMt4EjieWT1FfljMBJHI%2Fuploads%2Fgit-blob-123c5f914dd98cfce7bf6fd4cf748892514463f6%2FLive%20Session%20Pop-out.png?alt=media" alt=""><figcaption><p>The pop-out window</p></figcaption></figure>

1. **Voice**: The voice state with **Turn on** or **Turn off**. For a Samantha who only listens, her status instead.
2. **Shared screen**: What the meeting sees, with the one button that stops it: **Clear visual**, **Stop demo** or **Stop sharing**.
3. **Answer cards**: The same cards, compact. Suggested visuals are rows with **Show**, and a demo topic has **AI Demo**, with its visuals listed under "2 visuals".

* In Chrome, the pop-out is a picture-in-picture window that stays on top of other apps, including your meeting. Other browsers open a small popup window.
* If your browser blocks it, you see "Couldn't open the pop-out. Allow pop-ups for this site."
* It runs from the live session screen, so keep that tab open. Closing or reloading the tab closes the pop-out.
* When the call ends, the pop-out shows **Close**.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.spara.com/solutions-engineer/in-the-meeting/live-session-screen.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.
