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

# General Settings

> Configure your agent's name, AI model, instructions, tone of voice, and response behavior.

The General Settings page is where you define the core personality and behavior of your agent. This includes everything from the AI model it uses to how it speaks and how much detail it provides.

<Info>
  You can access General Settings by selecting an agent from the **Agents** list and clicking the **General** tab.
</Info>

## Agent name

Give your agent a clear, descriptive name that helps your team identify it at a glance. The name is visible in the RoseRx dashboard and in widget configurations.

* Maximum length: **255 characters**
* Choose something meaningful — for example, *"Aspirin Clinical Support"* or *"Cardiology Patient Navigator"*

## AI model selection

Choose from three model tiers, each balancing speed and capability differently. All tiers support a 200K-token context window.

| Tier         | Best for                                                                    |
| ------------ | --------------------------------------------------------------------------- |
| **Fast**     | Quick responses for high-volume use                                         |
| **Balanced** | The best balance of speed and quality — recommended and selected by default |
| **Advanced** | The most capable option for complex analysis and multi-step reasoning       |

<Tip>
  **Balanced** is recommended for most use cases. Choose **Advanced** for agents handling complex clinical queries, or **Fast** when response speed matters most.
</Tip>

## Agent instructions

The **Agent instructions** dropdown sets the agent type, which determines the core behavior of your agent:

* **HCP Clinical Decision Support** — Provides jurisdiction-aware clinical information, dosing guidelines, and evidence-based recommendations to healthcare professionals using professional medical terminology.
* **Patient Navigator** — Communicates in simple, accessible language, avoids medical advice, and follows strict compliance guidelines.
* **Medication Navigator** — Acts as a companion for patients on a prescribed medication, meeting them at their stage of the journey.
* **Field Team Co-Pilot** — Gives sales reps fast, compliance-safe talking points during HCP conversations.

## Focus fields

Depending on your agent type, you'll configure one or more focus fields that anchor the agent's responses:

<Tabs>
  <Tab title="HCP & Medication">
    ### Product

    Select the **product** the agent focuses on. The agent uses this to keep its responses anchored to the specific drug or therapeutic product, drawing from the knowledge base materials you've added.
  </Tab>

  <Tab title="Patient Navigator">
    ### Indicated conditions

    Specify the **indicated conditions** this agent covers. This helps the agent contextualize its responses within the relevant disease area — for example, *type 2 diabetes* or *chronic migraine*.
  </Tab>

  <Tab title="Field Team Co-Pilot">
    ### Product being promoted & competitors

    Specify the **product being promoted** and the **competitors** reps may be asked about. The agent only references competitors when the HCP raises them first, and always leads with your product's strengths.
  </Tab>
</Tabs>

## Additional instructions

Use additional instructions to give your agent custom behavioral guidance beyond its default instructions. This is useful for fine-tuning how the agent responds in specific scenarios.

To enable additional instructions:

1. Toggle the **Additional Instructions** switch to **on**
2. Enter your custom instructions in the text field

**Examples of additional instructions:**

* *"Always recommend patients speak with their doctor before changing medication"*
* *"When discussing dosing, always reference the approved prescribing information"*
* *"If the user asks about pricing, direct them to the website"*

<Tip>
  Keep additional instructions clear, specific, and concise. The agent follows these alongside its core instructions, so avoid contradicting the built-in behavior.
</Tip>

## Tone of voice

Set the conversational tone your agent uses when responding. The available options depend on the agent type.

<Tabs>
  <Tab title="Patient-facing agents">
    | Tone             | Description                                                             |
    | ---------------- | ----------------------------------------------------------------------- |
    | **Professional** | Warm but polished. Suitable for most patient-facing communications.     |
    | **Informal**     | Friendly and conversational. Ideal for consumer health applications.    |
    | **Enthusiastic** | Upbeat and encouraging. Works well for wellness and lifestyle products. |
  </Tab>

  <Tab title="HCP-facing agents">
    | Tone                     | Description                                                            |
    | ------------------------ | ---------------------------------------------------------------------- |
    | **Direct and Blunt**     | Concise, no-nonsense responses. Gets straight to the clinical detail.  |
    | **Measured and Neutral** | Balanced and objective. Presents information without emphasis or bias. |
  </Tab>
</Tabs>

## Response length

Control how much detail the agent provides using the response length sliders. Length is set **per channel**, so the same agent can give in-depth answers on the web while staying concise elsewhere. Each channel has four levels — **Minimal**, **Short**, **Long**, and **In-depth**.

For the **Web** and **Widget** channels, the levels target:

| Level        | Target length      | Best for                                       |
| ------------ | ------------------ | ---------------------------------------------- |
| **Minimal**  | \~500 characters   | Quick, focused answers to simple questions     |
| **Short**    | \~1,000 characters | Concise responses with essential detail        |
| **Long**     | \~2,000 characters | Thorough answers with supporting context       |
| **In-depth** | \~4,000 characters | Comprehensive responses with full explanations |

<Note>
  If your plan includes additional channels such as SMS, WhatsApp, or Email, a separate slider appears for each with channel-appropriate length targets. Actual response length may vary slightly depending on the complexity of the question.
</Note>

## Publishing your agent

Agents must be published before they can serve responses through widgets.

<Tabs>
  <Tab title="Draft agents">
    Draft agents show two buttons at the bottom of the settings page:

    * **Save** — Saves your changes without publishing. The agent remains in Draft status.
    * **Publish** — Saves your changes and sets the agent to **Active** status, making it available to connected widgets.
  </Tab>

  <Tab title="Active agents">
    Active agents show a **Publish Changes** button. When you modify settings on an active agent, changes are saved as a draft until you click **Publish Changes** to push them live.
  </Tab>
</Tabs>

<Tip>
  Always review your full agent configuration — including guardrails, knowledge base, and adverse event settings — before publishing for the first time.
</Tip>

## Danger zone

At the bottom of the General Settings page, the **Danger Zone** provides irreversible actions:

* **Archive agent** — Moves the agent to Archived status. It will be disconnected from all widgets and can no longer serve responses. Contact support if you need an archived agent reactivated.
* **Delete conversations** — Permanently removes all conversation history associated with this agent. This action is currently unavailable in the platform.

<Warning>
  Danger Zone actions are irreversible. Archiving disconnects the agent from all widgets. Proceed with caution.
</Warning>