> ## Documentation Index
> Fetch the complete documentation index at: https://docs.praxis-ai.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Meeting Agent

> Send your Digital Twin into a Zoom, Meet, or Teams call as a participant — with your memory, your files, and your vault travelling with it.

The Meeting Agent lets someone send their Digital Twin into a live video call. It joins as a
visible participant with a face and a voice, listens to the conversation, and — depending on how
you configure it — takes notes, suggests things privately, answers when addressed, or takes part
fully.

<Note>
  **The avatar in the meeting is that person's own twin.** Their memory, their files, and their
  vault travel with it. It is not a shared service account, and two people sending their twins into
  the same call each get their own.
</Note>

## Before you begin

Two things must be in place:

1. **Enable Anam Meeting Bot** on the instance (see below).
2. **An Anam avatar must be configured** for the Digital Twin — the meeting agent uses an avatar
   face and voice, so it cannot run without one. See
   [Realtime Voice & Avatars](/mdx/admin-guide/realtime-voice).

Settings live in the instance editor under **Configuration and Integrations → Integrations → Anam
Meeting Bot**.

## Sending the twin to a meeting

Once enabled, users ask their twin in conversation:

```
Pria, join https://zoom.us/j/123456789
```

The twin deploys an avatar into that call and reports back. Asking it to leave ends the session.
Both abilities switch on and off with the **Enable Anam Meeting Bot** setting — you do not manage
them separately.

<Warning>
  **Consent is your responsibility, not the platform's.** An avatar joining a call is a visible
  participant, and many jurisdictions and institutional policies treat an AI participant that
  listens to a meeting the same way they treat recording it. Tell participants, and follow your
  own policy on notice and consent, before rolling this out.
</Warning>

## Engagement mode

The single most important setting: how active the avatar is once it is in the room. You set the
instance default here; an individual deployment may narrow it.

| Mode                  | Behaviour                                                                                           |
| --------------------- | --------------------------------------------------------------------------------------------------- |
| **Silent scribe**     | Observes only. Never speaks, never posts. Still follows the conversation so it can recap afterwards |
| **Whisper copilot**   | Suggests privately to the person who sent it. Nothing reaches the meeting unless they act on it     |
| **Asked participant** | Speaks and posts when directly addressed, and otherwise stays quiet                                 |
| **Active delegate**   | Takes part fully, contributing when it judges it useful                                             |

<Tip>
  Start at **Silent scribe** or **Asked participant**. They are the two modes people find
  comfortable when an avatar is new to a team, and you can raise the level once expectations are
  set.
</Tip>

### Whisper proposals

Available with **Whisper copilot**. The avatar privately proposes chat messages to the person who
sent it, and **each one must be approved before anything is posted** to the meeting. Nothing is
posted on the agent's own initiative.

## Identity and voice

How the avatar appears and sounds in the call.

<AccordionGroup>
  <Accordion title="Anam Persona ID">
    The saved Anam persona the avatar wears — its **face and voice**. Leave it empty to fall back to
    the Digital Twin's own configured avatar and voice.

    <Note>Only the persona's appearance and voice are used. Its behaviour is always your Digital Twin's.</Note>
  </Accordion>

  <Accordion title="Meeting Bot Display Name">
    The participant name shown in the call — for example `RAI` or `Touchdown Timmy`. Empty defaults
    to `Pria`.
  </Accordion>

  <Accordion title="Wake-Word Aliases">
    Extra names the avatar answers to when spoken aloud, comma-separated. Speech recognition often
    mishears a stylized name — `RAI` is commonly heard as *Ray* — so listing the variants makes it
    respond reliably.

    <Warning>
      **Setting a display name or any alias drops the universal `Pria` wake word.** From then on the
      avatar answers only to its configured name and the aliases you list. If you rename it, add
      every variant people will actually say.
    </Warning>
  </Accordion>

  <Accordion title="Pronunciations">
    Comma-separated `written=spoken` pairs that fix how a stylized name is **pronounced** — for
    example `RAI=Ray, R-A-I=Ray`. List each written variant explicitly.

    This affects spoken output only. Text the avatar posts into meeting chat is never altered.
  </Accordion>

  <Accordion title="Say 'avatar' (never 'digital twin')">
    When on, the avatar describes itself as an *avatar* and never as a *digital twin*, in both what
    it says and what it writes. Use this where "digital twin" is the wrong register for your
    audience.

    Applies to meetings only — it does not change how the twin talks about itself in ordinary
    one-to-one conversations.
  </Accordion>
</AccordionGroup>

## Behaviour in the call

<AccordionGroup>
  <Accordion title="Maximum spoken answer length">
    A target ceiling, in seconds, on how long a single spoken answer runs. Accepted values are
    10–180 seconds; `0` means no limit.

    <Tip>Meetings punish long answers far more than chat does. A 20–30 second ceiling keeps the avatar conversational rather than lecturing.</Tip>
  </Accordion>

  <Accordion title="Default skill">
    An assistant the avatar loads on top of the Digital Twin's usual behaviour — useful when
    meetings need a consistent specialism.

    The meeting avatar does **not** inherit whichever assistant the person happens to have active in
    their own chat window, so if you want a particular one in meetings, set it here. Empty is a
    perfectly good default: the twin's own behaviour answers. Assistant IDs come from the
    [Assistants](/mdx/admin-guide/assistants) list.
  </Accordion>

  <Accordion title="Region">
    Which regional service the avatar deploys against. The region is fixed when the avatar joins,
    so changing this setting never moves an avatar that is already in a call — it applies to the
    next one.
  </Accordion>
</AccordionGroup>

## After the meeting

When the call ends, the twin writes up what happened, and what it learned can be saved to the
person's memory and their vault so it is available in later conversations.

Because the avatar was that person's own twin, the recap and anything remembered belong to **them**
— not to a shared pool, and not to other attendees who sent their own twins to the same call. What
each twin remembers is scoped to its own person, exactly as it is in ordinary conversation. See
[Twin Memory](/mdx/user-guide/memory/twin-memory).

## Costs

Meeting sessions consume credits like any other twin activity, and a meeting that runs for an hour
with an avatar listening throughout costs meaningfully more than a short chat. If the Digital Twin
pools credits, the pool pays; otherwise it comes from the individual's balance.

<Tip>
  Before rolling this out widely, set per-user quotas under **Configuration → User Quotas** so one
  enthusiastic pilot user cannot drain a shared pool. See [Credit Management](/mdx/admin-guide/credit-management).
</Tip>

## Related

* [Configuration and Integrations](/mdx/admin-guide/configuration#anam-meeting-bot) — where these settings live
* [Realtime Voice & Avatars](/mdx/admin-guide/realtime-voice) — avatar face, voice, and provider setup
* [Assistants](/mdx/admin-guide/assistants) — creating the skill the avatar can preload
* [Twin Memory](/mdx/user-guide/memory/twin-memory) — what the twin keeps afterwards
* [Credit Management](/mdx/admin-guide/credit-management) — pooled credits and per-user quotas