SQL Assessment API

Applies to: SQL Server 2012 (11.x) and later Azure SQL Managed Instance SQL Server on Azure Virtual Machines SQL Server on Linux

The SQL Assessment API provides a mechanism to evaluate the configuration of your SQL Server for best practices. The API is delivered with a ruleset containing best practice rules suggested by the SQL Server team. This ruleset is enhanced with the release of new versions, but at the same time, the API is built to give a highly customizable and extensible solution. Users can tune the default rules and create their own.

The SQL Assessment API is useful when you want to make sure your SQL Server configuration is in line with recommended best practices. After an initial assessment, configuration stability can be tracked by regularly scheduled assessments.

The API can be used to assess:

The API is also used by SQL Server Assessment Extension for Azure Data Studio (ADS).


The SQL Assessment API provides assessment on a variety of areas, but it does not go deeply into security. We recommend you use SQL Vulnerability Assessment to proactively improve your database security.


Rules (sometimes referred to as checks) are defined in JSON formatted files. The ruleset format requires a ruleset name and version to be specified. When you use custom rulesets, you can easily know which recommendations from what ruleset come.

The Microsoft's shipped ruleset is available on GitHub. You can view the entire ruleset in the samples repository.

SQL Assessment cmdlets and associated extensions

Use the API directly

The SQL Assessment API is available and can be used through managed code as part of any of these components:

Before you start using the SQL Assessment API itself, make sure to install either of these:

The SMO Framework is supplemented by the SQL Assessment API extension that provides the following methods:

  • GetAssessmentItems: Returns available checks for a particular SQL object (IEnumerable<...>)

  • GetAssessmentResults: Synchronously evaluates assessment and returns results and errors if any (IEnumerable<...>)

  • GetAssessmentResultsList: Asynchronously evaluates assessment and returns results and errors if any (Task<...>)

Use the API via PowerShell

If you would like to invoke the SQL Assessment API via PowerShell, you must install SQL Server PowerShell module. The SqlServer module provides two cmdlets that work with the SQL Assessment API:

  • Get-SqlAssessmentItem: Provides a list of available assessment checks for a SQL Server object

  • Invoke-SqlAssessment: Provides results of an assessment

Get started using SQL Assessment cmdlets

An assessment is performed against a chosen SQL Server object. In the default ruleset, there are checks for two kinds of objects only: Server and Database (in addition to them, the API supports two more kinds: Filegroup and AvailabilityGroup). If you want to assess a SQL Server instance and all its databases, you should run the SQL Assessment cmdlets for each object separately. Or, you can pass objects for assessment to the SQL Assessment cmdlets in a variable or the pipeline.

SqlServer and RegisteredServer objects are interchangeable, so you can pass any to the SQL Assessment cmdlets.

Go through the following examples to get started.

  1. Get a list of available checks for a local default instance to familiarize yourself with the checks. In this example, we're piping the output of the Get-SqlInstance cmdlet to the Get-SqlAssessmentItem cmdlet to pass the instance object to it.

    Get-SqlInstance -ServerInstance 'localhost' | Get-SqlAssessmentItem
  2. Get a list of available checks for all databases of the instance. Here, we're using the Get-Item cmdlet and a path implemented with the Windows PowerShell SQL Server provider to get a list of the databases, and then piping it to the Get-SqlDatabase cmdlet.

    Get-Item SQLSERVER:\SQL\localhost\default | Get-SqlAssessmentItem

    Also, you can use the Get-SqlDatabase cmdlet to do the same.

    Get-SqlDatabase -ServerInstance 'localhost' | Get-SqlAssessmentItem
  3. Invoke assessment for the instance and save the results to a SQL Server table. In this example, we're piping the output of the Get-SqlInstance cmdlet to the Invoke-SqlAssessment cmdlet, which results are piped to the Write-SqlTableData cmdlet. The Invoke-Assessment cmdlet is run with the -FlattenOutput parameter in this example. This parameter makes the output suitable for the Write-SqlTableData cmdlet. The latter raises an error if you omit the parameter.

    Get-SqlInstance -ServerInstance 'localhost' |
    Invoke-SqlAssessment -FlattenOutput |
    Write-SqlTableData -ServerInstance 'localhost' -DatabaseName SQLAssessmentDemo -SchemaName Assessment -TableName Results -Force

    Now let's invoke an assessment for all databases of the instance and add the results to the same table.

    Get-SqlDatabase -ServerInstance 'localhost' |
    Invoke-SqlAssessment -FlattenOutput |
    Write-SqlTableData -ServerInstance 'localhost' -DatabaseName SQLAssessmentDemo -SchemaName Assessment -TableName Results -Force
  4. Follow descriptions and links in the table to further understand the recommendations.

  5. Customize the rules based on your environment and organizational requirements (see below).

  6. Schedule a task or a job to run the assessment regularly or on-demand to measure progress.

Customize rules

Rules are designed to be customizable and extensible. Microsoft's ruleset is designed to work for most environments. However, it's impossible to have one ruleset that works for every single environment. Users can write their own JSON files and customize existing rules or add new ones. Examples of customization and complete Microsoft released ruleset are available in the samples repository. For more details on how to run the SQL Assessment cmdlets with custom JSON files, use the Get-Help cmdlet.

Options available with rule customization feature

Enable/disable certain rules or groups of rules (using tags)

You can silence specific rules when they aren't applied to your environment, or until scheduled work is done to rectify the issue.

Change threshold parameters

Specific rules have thresholds that are compared against the current value of a metric to find out an issue. If the default thresholds don't fit, you can change them.

Add more rules written by you or third parties

You can combine rulesets by adding one or more JSON files as parameters to your SQL Assessment API call. Your organization might write those files or obtain them from a third party. For example, you can have your JSON file that disables specific rules from the Microsoft ruleset, and another JSON file by an industry expert that include rules you find useful for your environment, followed by another JSON file that changes some threshold values in that JSON file.


We urge you not to use rulesets that come from untrusted sources, until you thoroughly review them to make sure they are safe.

Next steps