---
sourceDocument: Australia Platform security
sourceDocumentLink: https://servicenow-prod.fluidtopics.net/r/de-DE/platform-security

 Release :

    - australia

ft:locale :

    - de-DE

ft:publication_title :

    - Australia Platform security

ft:clusterId :

    - psec

bundleId :

    - psec

workflow :

    - Platform


---

# Enable Guardian for External Agents

# Enable Guardian for External Agents {#ariaid-title1}

* Freigeben Version: Australia
* 
* Aktualisiert 12. März 2026
* 
* ![](https://www.servicenow.com/docs/portal-asset/ico-clock) 1 Minute Lesedauer

Use a system property to protect your Language Learning Models (LLMs) with Guardian.
Use the sn_aia.external_agent_guardian_check system property to enable Guardian, which monitors requests sent to Language Learning Models (LLMs) and their responses to help protect you, your users, and your data. There are three types of content that are monitored for:

* offensive or harmful content
* prompt injection attempts
* filtered subjects
{#sc-enable-guardian-for-external-agents__ul_x5t_nxs_4hc}

Ensure that the sn_aia.external_agent_guardian_check system property does not exist in the System Properties \[sys_properties\] table, or exists and is set to a value of <kbd class="ph userinput">true</kbd>.

## More information {#sc-enable-guardian-for-external-agents__section_uhx_1xs_4hc}

{#sc-enable-guardian-for-external-agents__table_w2n_pqs_4hc__entry__2}

| Attribute | Description |
|-|-|
| Configuration name | sn_aia.external_agent_guardian_check |
| Configuration type | System Properties (/sys_properties_list.do) |
| Data type | Boolean |
| Recommended value | true |
| Default value | true |
| Fallback value | true |
| Category | [Access control](https://servicenow-prod.fluidtopics.net/xAmazkRh5FKyeb7q4TQGWw "The access control category audits the process of protecting resources from unauthorized access through granting and denying requests based on a permission model. This includes ensuring an entity accessing a resource holds valid credentials to do so, creating and protecting a well-defined set of roles or permissions and ensuring role or permission controls are protected from replay and tampering.") |
| Security risk | * Severity score:4.1 * CVSS score: Medium * Security risk details: Prompt injection is a type of attack where bad actors override the normal instructions of an LLM to access restricted information or elicit unexpected behaviors. Prompt injection detection is based on the LLM which has been trained on various types of prompt injection techniques such as role playing, paraphrasing, repetition, instructions to ignore other instructions, persuasion, etc. {#sc-enable-guardian-for-external-agents__ul_x2n_pqs_4hc} |
| Functional Impact | Guardian may produce false positives and block legitimate requests. |
| Dependencies and prerequisites | None |
[ ]

{#sc-enable-guardian-for-external-agents__table_w2n_pqs_4hc}