# Deploy Support

Deploy support enables you to version control and deploy your Umbraco.AI configuration across environments (development, staging, production).

## What Gets Deployed

With Deploy support installed, these AI entities are automatically deployed:

* **Connections** - Provider settings and connection configuration
* **Profiles** - Model selection and capability-specific settings (chat, embedding, speech-to-text)
* **Contexts** - Reusable context resources for AI operations
* **Guardrails** - Input/output safety and compliance rules
* **Settings** - Default profile assignments (chat, embedding, classifier)
* **Prompts** - Prompt templates and scoping (requires Umbraco.AI.Prompt.Deploy)
* **Agents** - Agent definitions, tool permissions, and workflows (requires Umbraco.AI.Agent.Deploy)

## How It Works

When you save an AI entity in the Umbraco backoffice, Deploy automatically:

1. Creates a deployment file in your repository (e.g., `data/Revision/umbraco-ai-connection__*.uda`)
2. Commits the file to version control alongside your code
3. Deploys the configuration when you push to other environments

Your AI configuration follows the same deployment workflow as your content types and other Umbraco configuration.

## Packages

| Package                      | Purpose                                                                           |
| ---------------------------- | --------------------------------------------------------------------------------- |
| **Umbraco.AI.Deploy**        | Core Deploy support for Connections, Profiles, Contexts, Guardrails, and Settings |
| **Umbraco.AI.Prompt.Deploy** | Deploy support for Prompt templates                                               |
| **Umbraco.AI.Agent.Deploy**  | Deploy support for AI Agents                                                      |

## Installation

Install the packages you need via NuGet:

{% code title="Terminal" %}

```bash
# Core (required)
dotnet add package Umbraco.AI.Deploy

# Optional add-ons
dotnet add package Umbraco.AI.Prompt.Deploy
dotnet add package Umbraco.AI.Agent.Deploy
```

{% endcode %}

See [Installation](/ai-in-umbraco/add-ons/deploy/installation.md) for detailed setup instructions.

## Key Features

### Sensitive Data Protection

Deploy automatically filters sensitive data like API keys to prevent secrets from being committed to version control.

Use configuration references (e.g., `$OpenAI:ApiKey`) instead of hardcoded values to keep secrets safe. See [Configuration](/ai-in-umbraco/add-ons/deploy/configuration.md) for details.

### Dependency Resolution

Deploy automatically resolves dependencies between entities:

* Profiles depend on Connections (and any Guardrails referenced in chat profiles)
* Prompts can optionally depend on Profiles, plus any referenced Guardrails
* Agents can optionally depend on Profiles, plus any referenced Guardrails
* Settings can optionally depend on the Profiles they reference

If you deploy a Profile, Deploy ensures its Connection is deployed first.

### Multi-Environment Support

Deploy the same AI configuration to multiple environments:

* **Development** - Use test API keys from configuration
* **Staging** - Use staging API keys from configuration
* **Production** - Use production API keys from configuration

Each environment resolves configuration references from its own `appsettings.json`.

## Next Steps

* [Installation](/ai-in-umbraco/add-ons/deploy/installation.md) - Install Deploy packages
* [Configuration](/ai-in-umbraco/add-ons/deploy/configuration.md) - Configure sensitive data filtering
* [Deploying Entities](/ai-in-umbraco/add-ons/deploy/deploying-entities.md) - Deploy AI configuration between environments
* [Best Practices](/ai-in-umbraco/add-ons/deploy/best-practices.md) - Security and workflow recommendations


---

# 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.umbraco.com/ai-in-umbraco/add-ons/deploy.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.