Access Azure Advisor recommendations for multiple subscriptions

Neha Bajpai (BAP) 0 Reputation points Microsoft Employee
2025-10-16T18:08:11.4366667+00:00

I am creating an AI agent at Organization Level that included multiple services under it having multiple subscriptions. Is there any Kusto cluster on which I can get read only access and filter all Azure Cost recommendations for multiple subscriptions?

Currently I will need to provide my managed identity to read access to all subscriptions to read the advisor recommendation.

Azure Advisor
Azure Advisor
An Azure personalized recommendation engine that helps users follow best practices to optimize Azure deployments.
{count} votes

1 answer

Sort by: Most helpful
  1. Sandhya Kommineni 1,575 Reputation points Microsoft External Staff Moderator
    2025-10-17T06:51:50.7+00:00

    Hello Neha Bajpai (BAP),

    Thanks for posting your question in Microsoft Q&A forum

    No, there is no centralized public Kusto cluster provided by Microsoft where you can query Azure Cost or Advisor recommendations across multiple subscriptions with just a read-only permission.

    Azure Advisor recommendations are typically accessed via the Azure portal, the Advisor REST API, or the Azure CLI/PowerShell. The existing cost data solutions in Azure primarily focus on cost and usage details, not the Advisor recommendations themselves, as a centralized Kusto data feed.

    Here are the best approach and solution for your AI agent to consolidate and filter cost recommendations from multiple subscriptions, along with the necessary permissions:

    You can view recommendations on the Advisor dashboard in Azure portal. For more information, see Azure Advisor portal basics.

    1. Azure Advisor helps you optimize and reduce your overall Azure spend by identifying idle and underutilized resources. You can get cost recommendations from the Cost tab on the Advisor dashboard.
    • Sign in to the Azure portal.
    • Search for and select Advisor from any page.
    • On the Advisor dashboard, select the Cost tab. User's image

    Refer document: Configure Azure Advisor recommendations view - Azure Advisor | Microsoft Learn

    1. The most direct and programmatic way to access Azure Advisor recommendations for multiple subscriptions is by using the Azure Advisor REST API. This approach allows you to iterate through your subscriptions and gather all the cost recommendations.

    Refer: Recommendations - REST API (Azure Advisor) | Microsoft Learn

    GET https://management.azure.com/{resourceUri}/providers/Microsoft.Advisor/recommendations/{recommendationId}?api-version=2025-01-01
    

    The API allows filtering. You would need to make calls for each subscription and then consolidate and filter the results on your agent's side.

    1. You can use the Cost Management Query API to programmatically retrieve usage and cost data.

    Alternatively, if you prefer automated data collection, use the Cost Management Exports API to schedule daily exports of cost data to a storage account, which you can then ingest for analysis.

    POST https://management.azure.com/{scope}/providers/Microsoft.CostManagement/query?api-version=2025-03-01
    

    Refer document: Tutorial - Reduce Azure costs with recommendations - Microsoft Cost Management | Microsoft Learn

    Your current approach of granting your managed identity Reader access at the subscription level across all required subscriptions is correct and follows best practices for an organization-wide agent.

    The managed identity must have the Reader role assigned at the Subscription scope for each subscription from which it needs to retrieve recommendations.

    Refer document: Roles and permissions - Azure Advisor | Microsoft Learn

    I hope the provided answer is helpful, do let me know if you have any further questions on this Please accept as Yes and upvote if the answer is helpful so that it can help others in the community.

    1 person found this answer helpful.
    0 comments No comments

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.