---
sourceDocument: Australia Enable AI
sourceDocumentLink: https://servicenow-prod.fluidtopics.net/r/pt-BR/intelligent-experiences

 Release :

    - australia

ft:locale :

    - pt-BR

ft:publication_title :

    - Australia Enable AI

ft:clusterId :

    - platai

bundleId :

    - platai

workflow :

    - Platform


---

# Test user access

# Test user access to an AI agent {#ariaid-title1}

* Versão de lançamento: Australia
* 
* Atualizado 12 de mar. de 2026
* 
* ![](https://www.servicenow.com/docs/portal-asset/ico-clock) 2 min. de leitura

Run a manual test to verify that only the users you want to access the AI agent can do so.

## Antes de Iniciar

Role required: sn_aia_admin and either admin or at least one role required by the ACL of the AI agent being testing and each of the ACLs of its downstream components

## Por Que e Quando Desempenhar Esta Tarefa

You can establish the security settings for an AI agent in the guided setup to establish which users can access it. See [Define security controls](https://servicenow-prod.fluidtopics.net/YI7d7AFYLQVNMUUqkRTn3A "In the guided setup for an AI agent, define security controls for who can access the AI agent and what data the AI agent has access to.") for instructions on how to change the user access settings. When you select Save and continue on that step of the guided setup, an ACL is created that establishes limitations on which users can access the AI agent.

Once you have created these ACLs, you can verify that they work as intended by using the Test access test type of a manual test of an AI agent.

To see instructions for performing manual tests to evaluate performance, see [Test performance manually](https://servicenow-prod.fluidtopics.net/cKbj82uA3k2zRtzDBZ4jUg "Analyze the execution of an AI agent so you can see that it functions the way that you defined it."). For more information about automated tests, see [Evaluate an AI agent](https://servicenow-prod.fluidtopics.net/he2kFV6WLQwi6a4v1pckPQ "Evaluate agentic workflow and AI agents against datasets of your choice to monitor performance and evaluate against different benchmarks.").

## Procedimento

1. Navigate to AllAI Agent StudioTesting.
2. Select Start manual test.
3. In the Test type drop-down menu, select Test access.  
   If you want to test the basics of how an AI agent works, select AI agent or worfklow. The full instructions for that test type can be found in [Test performance manually](https://servicenow-prod.fluidtopics.net/cKbj82uA3k2zRtzDBZ4jUg "Analyze the execution of an AI agent so you can see that it functions the way that you defined it.").
4. Search for or select the name of the AI agent you want to test.
5. Select an invoking user.  
   The invoking user can be the user that triggers the AI agent or it can be the invoking user of an upstream component, such as an agentic workflow. For more information about how the invoking user works, see [Security for AI agents](https://servicenow-prod.fluidtopics.net/22j7AP47_LhOTCRi_NkWGw "Implement security controls for AI agents and agentic workflows through access control lists (ACLs) and user identities to increase alignment with the access control-based security measures in the agentic system.").

   When you select an invoking user, the user roles are populated in the Invoking user roles field. The field is read-only. If you want to change a user's roles, you must change the user's User record. See [Assign a role to a user](https://www.servicenow.com/docs/access?context=t_AssignARoleToAUser&version=australia&pubname=australia-platform-administration&ft:locale=en-US).
6. Select Test access.  

## Resultado

The results of the access test are opened in the Access Analyzer in a new browser tab.

Access Analyzer identifies all the ACL calls made in the execution of the AI agent, including its tools, if any are present. You can see the results of each ACL to verify that each one has the correct user access defined.

## O que Fazer Depois

If the results are different than what you expect or want, you can [redefine the security controls](https://servicenow-prod.fluidtopics.net/YI7d7AFYLQVNMUUqkRTn3A "In the guided setup for an AI agent, define security controls for who can access the AI agent and what data the AI agent has access to.") of the AI agent.

You can also test an AI agent's performance with either a manual test or automated evaluations. See [Test performance manually](https://servicenow-prod.fluidtopics.net/cKbj82uA3k2zRtzDBZ4jUg "Analyze the execution of an AI agent so you can see that it functions the way that you defined it.") or [Evaluate an AI agent](https://servicenow-prod.fluidtopics.net/he2kFV6WLQwi6a4v1pckPQ "Evaluate agentic workflow and AI agents against datasets of your choice to monitor performance and evaluate against different benchmarks.").