For the complete documentation index, see llms.txt. This page is also available as Markdown.
Cortex AgentiX

Fields and Types permissions

Control access to Cortex AgentiX custom fields and indicator types.

Controls access to custom fields and indicator types within Object Setup Settings → Configurations → Object Setup:

  • Case fields (Cases → Fields): Custom fields that extend the case data schema, appearing in case views, queries, and layouts.

  • Issue fields (Issues → Fields): Custom fields for the issue data schema, often used for automation rules and filtering.

  • Indicator fields and types: Definitions for custom indicator fields and new indicator types (e.g., Cloud Resource ID), including extraction regex patterns.

Permission
Description
Roles Example

None

No access to define fields or types. Users can still view and use existing fields in case/issue views, but cannot modify their definitions

SOC Tier 1 Analyst: Schema changes are outside Tier-1 scope; they use existing fields but don't need to see field configuration.

View

Read-only access to all field definitions and indicator types. Allows exporting definitions to CSV.

  • SOC Tier 2 and 3 Analysts: Need to understand field definitions for advanced queries, custom field usage, and investigation workflows.

  • Threat Hunter: Needs to understand field definitions for hunting queries (XQL) and custom field usage.

View/Edit

Full read/write access. Users can create, modify, or delete custom fields and indicator types, and write extraction regex.

Required and recommended permissions

As schema changes impact how data is displayed and analyzed, consider the following dependencies:

Permission
Permission Level
Reasons

Cases & Issues

View

Required to see how fields are used in actual cases/issues.

Layouts

View

Strongly recommended. See how fields are rendered in layouts; needed to design layouts that use custom fields.

Case Properties

View

Strongly recommended. Understand incident structure (statuses, domains) alongside field definitions

Threat Intel

View

Recommended. Understand indicator types and fields in the context of threat intelligence.

Audit

View

Recommended to track field creation/modification history.

Marketplace

View

Recommended to install content packs that include field definitions and indicator types.

Last updated

Was this helpful?