> For the complete documentation index, see [llms.txt](https://helpdocs.conducttr.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://helpdocs.conducttr.com/feature-documentation/designing-and-running/channels/monitoring-and-alerting-channels/social-watch/communications-manager.md).

# Communications Manager

The Communications Manager (or Comms Manager) is your **central hub for managing multiple role-playing personas** during an exercise. Instead of switching between channels, you can monitor and respond to all persona communications from one place.

It's especially useful in exercises role-players need to impersonate multiple personas. For personas with a full cognitive model, role-players can also set a persona **to auto-respond using AI.** Depending on the goals and experience of your training audience, auto-response might be best used for secondary or less crucial personas - freeing human role-players to manage leading personas.

<table><thead><tr><th width="149.333251953125">Benefit</th><th>Feature</th></tr></thead><tbody><tr><td>Easier to manage multiple impersonations</td><td><ul><li>See all incoming and outgoing messages across assigned personas</li></ul><p></p><ul><li>Reply manually as a persona, or let AI handle responses automatically</li></ul><p></p><ul><li>Monitor multiple social channels (MicroBlog [Twitter/X], GoSocial [Facebook], InstaMedia [Instagram], Telegraph [Telegram], Website/News [Browser], Forum [Reddit])</li></ul></td></tr></tbody></table>

### How does it work?

#### Behavioural Simulation Engine (Conducttr's 4Rs Model)

Comms Manager uses Conducttr's 4Rs model to power persona decision-making:

* Is the request **RELEVANT**?
* Does it **RESONATE**?
* How should I **REACT**?
* What's my **RESPONSE**?

This pipeline guides how personas process incoming messages, ensuring their responses stay true to their cognitive model and in-character persona.

<figure><img src="/files/Q1JLQABG4bj4hBOF552U" alt=""><figcaption></figcaption></figure>

## Designing

:zap:Before running the Comms Manager in Social Watch, work through this setup checklist:

* #### Activate the Social Watch channel and ensure that *Autonomous Personas (Auto)* is toggled on

<figure><img src="/files/p3ZwQ0quwNVWOKIa8AMs" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/xiBW4giLraG9g8Ccv4j4" alt=""><figcaption></figcaption></figure>

* #### To get the most credible AI-generated responses, complete the scenario description and ensure it provides adequate exercise context.  Personas must have full cognitive models

<figure><img src="/files/q4Tt0JGI6KlNmW9D5Jav" alt=""><figcaption></figcaption></figure>

<figure><img src="/files/R2uq7pA1xMGkV4Ikx0Zz" alt=""><figcaption></figcaption></figure>

{% hint style="info" %}
To find out more about persona attributes, visit the page below:
{% endhint %}

{% content-ref url="/pages/wRYkXCAB5l7N3Qxg2pTW" %}
[Working with personas](/feature-documentation/designing-and-running/personas/working-with-personas.md)
{% endcontent-ref %}

* #### Persona permissions and player positions need to be aligned for the human players to gain access to the persona

<figure><img src="/files/T5PF9gxrHFEZ41jp5uZK" alt=""><figcaption><p>Set up positions in <code>Configuration</code> > <code>Positions</code></p></figcaption></figure>

<figure><img src="/files/BoCUdy9ysFYFuFQoWjHk" alt=""><figcaption><p>You can also assign positions to personas directly</p></figcaption></figure>

<figure><img src="/files/WH5Oq9ZmFkW3YLedXgmt" alt=""><figcaption><p>Pre-assign positions to players in your teams table</p></figcaption></figure>

<figure><img src="/files/aLDVSXLwHtbCwrip91AE" alt=""><figcaption><p>Assign positions during a live exercise from the Teams tab of your Facilitator Dashboard</p></figcaption></figure>

* #### Set up Social Graph (social media following relationships between personas) which affects message visibility

<figure><img src="/files/Fm4osijJh6F6Dui1beGA" alt=""><figcaption><p><em>Example Social Graph manipulated in Excel</em></p></figcaption></figure>

{% hint style="info" %}
To find out more about Social Graph, visit the page below:
{% endhint %}

{% content-ref url="/pages/Ks7MAXCSFFuc1jXbopd3" %}
[Social Graph](/feature-documentation/designing-and-running/personas/working-with-personas/social-graph.md)
{% endcontent-ref %}

{% hint style="danger" %}
If any of these steps are incomplete, players may not see the right personas or messages during the exercise.
{% endhint %}

***

## Running

### Getting started

When you first open Social Watch in an exercise you'll be met with this *Get Started* window:

<figure><img src="/files/cYkFNJmky3o7zv8rXFHU" alt=""><figcaption></figcaption></figure>

To use the Comms Manager, select `Impersonate Persona`:

<figure><img src="/files/kbqtVlTq3flfboNZjl3v" alt=""><figcaption></figcaption></figure>

### Setting up your column

1. ***Select Sender***: choose the persona you want to impersonate from the dropdown. You'll only see personas Facilitators/designers have assigned to you
2. **Filter by channels** (optional): limit monitoring to specific channels
3. **Choose your response mode** (see below)

To add another column, click `+ Add column` and repeat the process.

<figure><img src="/files/U8Fd9E4G3RasvISf9ckt" alt=""><figcaption></figcaption></figure>

### Response modes

#### 1. Impersonation (manual)

Monitor all incoming messages and replies yourself.

Player messages are highlighted with a green *Player* tag so they're easy to spot. Click `Reply` to open the Composer and type your response, or use the AI sparkles button to generate one based on context you describe.

<figure><img src="/files/0Qa7ZtgnJUrc4cvgxlJH" alt=""><figcaption></figcaption></figure>

#### 2. Auto respond

AI automatically replies on behalf of the persona based on their cognitive profile.

This is useful when managing several personas at once. You can still see everything that's sent.

<figure><img src="/files/2xZNqOwuFIDsPm1qu3ZN" alt=""><figcaption></figcaption></figure>

{% hint style="warning" %}
Auto respond is currently available for MicroBlog (X/Twitter) and Email only. You can only select one channel per persona per column when using this mode.
{% endhint %}

<figure><img src="/files/uyhfXWu5nCV0FhGTixRg" alt=""><figcaption><p>An example of an Autonomous Persona in action</p></figcaption></figure>

### Replying as a persona

Click `Reply` on any message to open the Composer. From here you can:

* Type a response manually
* Click `Generate Response` to get an AI-drafted reply (you'll need to briefly describe the situation/context)

{% hint style="success" %}
You can address replies to any player or persona using the *To* and *CC* fields.
{% endhint %}

<figure><img src="/files/jeTIfibB5eHdP0qQFfqp" alt=""><figcaption></figcaption></figure>

### Useful things to know

#### MIND state

Navigate to the :pencil2: pencil icon for the *Composer* > :book: `Mind`:

<figure><img src="/files/sJHPKaCSQZiIDeWaZlCk" alt=""><figcaption></figcaption></figure>

Each persona has an emotional state that influences how they respond. It updates throughout the exercise as interactions build up.

Here you'll also be able to see *Relationship*, *Memory summary*, *MIND state*, and *FACTS*.

#### FACTS

Concrete information (road closures, policy rules, operational updates) can be added to a persona's knowledge base mid-exercise to keep AI responses accurate.

<div align="left"><figure><img src="/files/xC84s6dRtztypMJJPDu9" alt=""><figcaption></figcaption></figure></div>

#### Relationships

Click `Relationships` to view the relationship history between personas:

<div align="left"><figure><img src="/files/nx3DUdbbev9pkBbV03eC" alt=""><figcaption></figcaption></figure></div>

Comms Manager tracks how personas and players interact over time, shaping the tone of AI-generated responses.&#x20;

#### Exercise context

Navigate to the cog icon:

<figure><img src="/files/pb40NmL75fnJ15K9YfkJ" alt=""><figcaption></figcaption></figure>

If the scenario shifts during the exercise, you can update the AI's context, without needing to back into the Editor.

<figure><img src="/files/rHRq9ubYnFkzN0VkpDAY" alt=""><figcaption></figcaption></figure>


---

# 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 dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://helpdocs.conducttr.com/feature-documentation/designing-and-running/channels/monitoring-and-alerting-channels/social-watch/communications-manager.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

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.