Introduction

Sana AI's integration with Azure Blob Storage enables users to efficiently access, manage, and retrieve data stored in Azure, allowing for seamless information retrieval and knowledge management within an organization.

Integration Capabilities

This integration allows users to perform operations such as setting and retrieving metadata on Azure Blob objects, which enhances the accuracy and relevance of search results. Users can manage metadata directly within Azure Blob Storage, utilizing advanced AI capabilities to gain insights and summarize documents based on stored data.

Type of integration

This is a shared integration, permitting multiple users within the organization to access and utilize the integrated features collectively.

Scope and permissions

The integration accesses Blob, Container, and Object resources within Azure Blob Storage. It requires permissions to read and list data to ensure accurate information retrieval and metadata management. This access is crucial for performing operations like setting metadata headers and retrieving metadata without altering the content of the resources.

Integration setup

Access integration: In Sana, navigate to the integrations tab and select Azure Cloud Storage.

Click on Connect shared

Fill in the required information:

Azure Storage account: Enter the storage account name.
Source container: Provide the container ID.
Prefix: Specify a folder to index or leave blank to index the entire container.
SAS token: Create a SAS token at the container level with the following settings:
- Allowed services: Include Blob.
- Allowed resource types: Select both Container and Object.
- Allowed permissions: Include Read and List.
- Default expiration: Set to 8 hours or a reasonable time for your needs.
- Allowed protocols: Use HTTPS only.
- If using a custom domain, format the token as:

https://${storageAccount}.${domain}/${container}${SAS-token}

You can find it under Blob SAS URL (image).

Share with should be filled in with the scope of who should have access, choose a collection or the entire workspace.

Meta Data Fields

Purpose: Customize properties for display, search, and retrieval to enhance accuracy and relevance.
Standard fields:
- URL: The link to the resource.
- displayName: The name displayed for the resource.
- createdAt: The creation date, formatted as an ISO 8601 date string.
- previewImageUrl: A URL to a preview image associated with the resource.
Custom fields:
- Fields prefixed with customAttributes_ are used for enhancing retrieval. They allow additional metadata to improve search results. Only strings and numerical values are valid.
Setting Metadata on Azure Blob objects:
- Format: Metadata headers are name/value pairs formatted as x-ms-meta-name:string-value. Names are case-insensitive but preserve the original case.
- Operations: Metadata can be retrieved or set using GET, HEAD, and PUT operations. Setting metadata overwrites existing values.
- URI syntax:
  - Retrieve/set container metadata: GET/HEAD/PUT https://myaccount.blob.core.windows.net/mycontainer?comp=metadata&restype=container
  - Retrieve/set blob metadata: GET/HEAD/PUT https://myaccount.blob.core.windows.net/mycontainer/myblob?comp=metadata
- Standard HTTP properties: Include ETag, Last-Modified, Content-Type, etc., in addition to custom metadata.

Support and Contact information

For further information about Sana AI, please contact [email protected] via email.

Data handling & privacy

Sana AI is fully committed to data security and privacy. All data accessed by Sana AI is encrypted both in transit and at rest. Sana does not train any underlying language models on your data, ensuring the privacy of your information. Furthermore, Sana AI respects the underlying permissions of SharePoint with the individual integration, ensuring that users can only access data they are authorized to view. Sana AI is ISO 27001 certified; and SOC 2 and GDPR compliant, and adheres to the highest standards of data security.