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

# Nova

> A helper in the console that looks things up and suggests changes. You apply them.

Ask Nova to set Xenovia up with you, explain what changed, or draft a policy. Nova looks things up and suggests changes; a person on your team applies every one.

For example, ask "What has this policy been deciding, and is it set up the way it should be?" on a policy's page, and Nova reads its recent decisions and answers. If it suggests a change, the change waits for you.

> [!NOTE]
>
> Nova appears only when it is enabled for your Xenovia deployment. When it's off, the **Nova** item, **Ask Nova** and every **with Nova** button are hidden. Ask your Xenovia contact or administrator to turn it on.

## Where to find it

* **Nova** at the top of the sidebar opens Nova's home: what it found that needs fixing, its suggestions, and what it is running.
* **Ask Nova**, at the bottom right of every page, opens a chat about the page you're on. Turn on **Include this page** to let Nova see what you're looking at. The chat can float, sit in a side panel or go full screen.
* The search box at the top (`⌘K` or `Ctrl+K`) can **Ask Nova** too.

![Nova panel answering “Why is this failing?” with a summary of a blocked policy, an explanation and suggested follow-ups](https://mintcdn.com/xenovia/RO5UF1Tk8L9B7yPt/images/screens/nova-panel.png?fit=max\&auto=format\&n=RO5UF1Tk8L9B7yPt\&q=85\&s=e21ca5b51b5c4fa1faa1da001e14d07e)

The Nova panel explaining why calls were blocked

## What Nova does

| Where | What it does |
| - | - |
| First run | **Set it up with Nova**, in the Welcome dialog and the **Get started** checklist, asks which coding agents your team uses and whether your code is on GitHub, then does the steps with you |
| Agents | **Set up with Nova**: describe the agent and Nova fills in the setup form |
| Policies, guardrails, capabilities, alert rules | **Create with Nova**: Nova asks questions one at a time and fills in the form. Each has a **Create manually** option beside it |
| Discovery | **Plan the setup**: Nova reads an agent's code and writes what Xenovia should set up for it. **Edit with Nova** refines the plan |
| Simulation | **Generate with Nova** writes test scenarios from an agent's sessions and code, as drafts for you to review. See [Simulation](https://docs.xenovia.io/simulation.md) |
| Pages | **Ask Nova** on a policy, an alert, a compliance control or a simulation run starts a chat about it |

## You approve every change

Nova never changes anything by itself.

* **Suggestions.** A change Nova proposes shows as a card marked **Suggested**, with **Apply** and **Discard**. Only admins can apply; members see the change. A suggestion whose target was edited since, or that expired, can't be applied: ask Nova for a fresh one.
* **Forms.** When Nova creates something, it fills in the product's own form, marking what it chose. You finish with the form's own **Create**.
* **Credentials.** Keys and credentials always go through the product's own forms. To connect an integration, applying the suggestion takes you to the integration's page, where you enter the credentials yourself. A message that looks like it holds a key is held with a warning before it reaches Nova.

When a fix is made, Nova checks that finding again. **Ignore** on a finding tells Nova not to raise it again.

## Credits

Nova uses your organization's AI credits. When they run out, Nova answers again once credits are added.

## What next?

> [card] [Quickstart](https://docs.xenovia.io/quickstart.md)
>
> Set up Xenovia, with or without Nova.

> [card] [Policies](https://docs.xenovia.io/control/policies.md)
>
> Create a policy, manually or with Nova.

> [card] [Simulation](https://docs.xenovia.io/simulation.md)
>
> Let Nova write scenarios to test your agent.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.