# Cobrowse virtual agents

Cobrowse provides ready-to-use virtual agents powered by ElevenLabs, offering seamless chat and voice interactions with your users.

## Enablement

Virtual agents are configured in the settings panel of the Cobrowse dashboard. To enable production-grade virtual agents, add your [ElevenLabs](https://elevenlabs.io) API key.

Each virtual agent has:

* A **display name** that users see (e.g., "Jessica")
* An **identifier** used for SDK integration

## SDK integration

Configure the virtual agent before initializing Cobrowse:

```javascript
CobrowseIO.virtualAgent = 'my-agent-identifier';
```

When Cobrowse starts, the chat and voice widget will automatically appear on the page.

## Trial agents

A free trial agent is available for testing. Trial agents have limited conversations, duration, and concurrency.

By default, trial agents perform web searches at the start of each conversation to retrieve relevant help information for users.

## Customizing production agents

Production virtual agents connected to your ElevenLabs account have no usage limits and offer [full customization options](https://elevenlabs.io/docs/agents-platform/overview), including:

* Telephony support
* Conversation behavior and personality
* Voice selection and tuning
* Widget appearance and branding
* Additional MCP integrations

**Note:** Production agents do not perform automatic web searches. You must configure a [knowledge base](https://elevenlabs.io/docs/agents-platform/customization/knowledge-base/rag) with relevant content to provide context for your virtual agent.

## Billing

Virtual agent sessions are billed separately from standard Cobrowse usage.

{% hint style="success" %}
Have questions about virtual agent pricing? Contact us at <hello@cobrowse.io>.
{% endhint %}


---

# Agent Instructions: 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:

```
GET https://docs.cobrowse.io/ai/cobrowse-agents.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.