Share via

Add Azure AI Search as a knowledge source

Azure AI Search provides a powerful search engine that can search through a large collection of documents. Copilot Studio supports adding Azure AI Search as a knowledge source.

To complete the connection, you need an Azure account. If you don't have an Azure account, you can create an account at Microsoft Azure.

After you create the Azure account, use the following Azure articles to set up and configure Azure AI Search. These articles provide information on the setup and configuration needed to use the Azure AI Search connection as a knowledge source:

Copilot Studio supports vectorized indexes using integrated vectorization. Prepare your data and choose an embedded model, then use Import and vectorize data in Azure AI Search to create vector indexes. This method enables the system to use the same embedded model used to vectorize the data to also vectorize the incoming prompt at runtime, which reduces the need to write special functions to do the same.

Copilot Studio also supports the use of the semantic ranker feature. This feature also needs to be configured in Azure AI Search before adding the connection in Copilot Studio. For more information, see How to get started with semantic ranker.

  1. Open the agent.

  2. Select Add knowledge from either the Overview or Knowledge pages, or the Properties of a generative answers node.

  3. From the Add knowledge dialog, select Featured.

  4. Select Azure AI Search.

  5. Select Create new connection.

  6. Select the Authentication type:

    • Access Key
    • Client Certificate Auth
    • Service principal (Microsoft Entra ID application)
    • Microsoft Entra ID Integrated

  7. Type the Azure AI Search Endpoint URL and the Azure AI Search Admin Key.

  8. Select Create again to complete the connection. A green check mark appears to confirm the connection.

  9. Select Next.

  10. Enter the Azure AI Search vector index to be used. Only one vector index can be added.

  11. Select Add to complete the connection.

After you add the connection, it appears in the knowledge sources table. The status displays as In progress while Copilot Studio indexes the metadata in the tables. After the indexing is complete, the status is updated to Ready, and then you can begin testing the knowledge source. During testing, you can verify that proper references were called by reviewing the files and citations cited by the agent.

Return citations

To return citations when using Azure AI Search in Copilot Studio, include a URL field with the actual link to the document in the search index. When the metadata_storage_path field is included in the index, Copilot Studio interprets that field as the citation. However, if that field doesn't exist, Copilot Studio considers whichever field that contains a complete URL link as the citation. For more information, see Index file content and metadata by using Azure AI Search

Note

When configuring citations in Azure AI Search, ensure that the users of your agent have the necessary permissions to access the data sources the citations point to. For example, if you add a URL in the search index that points back to a website, or knowledge base, the users should have access to those sources. If the URL points to a restricted data source, the users can't access the generated citations.

Virtual Network support

Copilot Studio supports Azure AI Search indexes configured for virtual networks. This configuration allows you to use a private endpoint to securely connect your search indexes.

For instructions on how to set up the endpoint, see Create a Private Endpoint for Azure AI Search

Configure Virtual Network support in Power Platform environment, see Set up Virtual Network support for Power Platform.

In Copilot Studio, follow the steps to complete the configuration of the connection.