> For the complete documentation index, see [llms.txt](https://cortex-docs.paloaltonetworks.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://cortex-docs.paloaltonetworks.com/cortex-agentix/configure-cortex-agentix/users-and-roles-management/roles-management/role-permissions-by-component/cortex-agentic-assistant-permissions/cortex-agentic-assistant-agents.md).

# Cortex Agentic Assistant Agents

Controls whether a user can access and interact with the Cortex Agentic Assistant, including managing actions and agents. This is the base permission required for any assistant interaction.

| Permission | Description                                                                                                                                                                                                                                                                                                             | Roles Example                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                           |
| ---------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| None       | The assistant panel is disabled. Users cannot use natural language queries or access the Agentic Assistant Hub.                                                                                                                                                                                                         |                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                         |
| View/Edit  | <p>Users can open the Cortex Agentic Assistant panel, access the Agentic Assistant Hub, view chat history, and view agent details and action details in the Hub. In addition, you can select the following:</p><ul><li>Interact with Agents</li><li>Manage Actions</li><li>Manage Agents</li><li>Agents Admin</li></ul> | <ul><li>SOC Tier-1 Analyst: Needs to use the assistant for quick lookups, IP/hash enrichment, and guided investigation. No need to manage agents or actions.</li><li>SOC Tier-2 and 3 Analysts: full assistant interaction for complex investigations. No need to manage agents or actions.</li><li>Threat Hunter: Full assistant interaction for threat hunting. Can view agents and actions, but does not need to modify them</li><li>Security Engineer: Builds custom agents and actions for the organization. Does not need Agents Admin as they manage their own agents.</li></ul> |

Cortex Agentic Assistant Agents sub-permissions

| Sub-permissions      | Description                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                   | Roles Example                                        |
| -------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------- |
| Interact with Agents | <p>Interact with Agents: Users can trigger Agents in the Cortex Agentic Assistant. Users can access their own agents, public agents, and system agents. For more information, see <a href="/pages/dlRVaVqH4FY8kyLca8yH">Chat with an Agentic Assistant agent</a>.</p><ul><li>Checked: The user can take actions, such as sending messages and queries to the assistant, starting new conversations with agents, and executing insight actions (Add as IOC, etc.).</li><li>Unchecked: The user can view the assistant panel, home screen, Agentic Assistant Hub (read-only, based on the parent View/Edit level), chat history, and previous conversations. The user can also view insights cards for IPs, hashes, and domains.</li></ul>                                                                                                                                                                                                                                                      | All roles should require full assistant interaction. |
| Manage Actions       | <p>Controls the ability to manage agent actions - the discrete operations that agents can perform. Actions are the building blocks that agents use to execute tasks. For more information, see <a href="/pages/H3BVVgDFFS8OfuTPNiky">Manage actions</a>.</p><ul><li>Checked: Users can manage actions, register scripts as agent actions, and configure action parameters and descriptions.</li><li>Unchecked: The user can view the Actions tab in the Agentic Assistant Hub (read-only, browse available actions and their configurations. They can also view action details, parameters, and descriptions.</li></ul>                                                                                                                                                                                                                                                                                                                                                                       | Security Engineer                                    |
| Manage Agents        | <p>Controls the ability to create and manage AI agents. Users with this permission can create custom agents, edit their own agents, and configure agent properties. For more information, see <a href="/pages/HnnADm4YDcXIRsyGHv3b">Manage agents</a>.</p><ul><li>Checked: Users can manage agents, configure agent instructions, and assign/remove actions from agents.</li><li>Unchecked: Users can view the Agents tab in the Agentic Assistant Hub, browse available agents, and their configurations. They can also view agent details, descriptions, and assigned actions.</li></ul>                                                                                                                                                                                                                                                                                                                                                                                                    | Security Engineer                                    |
| Agents Admin         | <p>The highest-level agent management permission. When checked, it overrides the ownership restrictions of the Manage Agents checkbox, allowing the user to edit and manage ALL agents in the organization, including system agents and agents created by other users. For more information, see <a href="/pages/2x7GoJ5LM4cGQxFVj0H1">Agentic Assistant Hub</a>.</p><ul><li>Checked: Users can edit/delete any agent, including system agents, configure custom instructions for system agents, and manage organization-wide agent settings. The user can override all agent ownership restrictions.</li><li>Unchecked: The user cannot edit system agents, cannot edit agents created by other users, and cannot manage organization-wide agent configurations.</li></ul><div data-gb-custom-block data-tag="hint" data-style="info" class="hint hint-info"><p><strong>Note</strong></p><p>If <strong>Manage Agents</strong> is checked, the user can only edit their own agents.</p></div> |                                                      |

**Required and recommended permissions**

Consider adding the following permissions:

| Permission                                    | Permission Level  | Reason                                                                                                                                                                                                                                   |
| --------------------------------------------- | ----------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| Cases & Issues                                | View or View/Edit | <ul><li>View: Required. Agent conversations reference cases and issues.</li><li>View/Edit: Strongly recommended. Agents can modify case status and assign analysts.</li></ul>                                                            |
| Threat Intelligence (under Threat Management) | View or View/Edit | <ul><li>View: Required. Agents perform Indicator lookups and enrichment.</li><li>View/Edit: Strongly recommended for Indicator actions. Agents can create/modify Indicators.</li></ul>                                                   |
| Query Center                                  | View or View/Edit | <ul><li>View: Required. Agents generate and execute XQL queries.</li><li>View/Edit: Strongly recommended. Agents execute queries on behalf of users.</li></ul>                                                                           |
| Scripts & Playbooks                           | View or View/Edit | <ul><li>View: Required. Register scripts as agent actions. Agents can suggest and reference playbooks.</li><li>View/Edit: Strongly recommended. Create scripts to register as actions; needed for full action builder workflow</li></ul> |
| Dashboards                                    | Enabled           | Strongly recommended to view the Command Center dashboard with agent integration.                                                                                                                                                        |
| Marketplace                                   | View/Edit         | Recommended to install content packs with agent definitions and actions.                                                                                                                                                                 |
| Integrations                                  | View/Edit         | Recommended to configure integrations that agents use as data sources.                                                                                                                                                                   |


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://cortex-docs.paloaltonetworks.com/cortex-agentix/configure-cortex-agentix/users-and-roles-management/roles-management/role-permissions-by-component/cortex-agentic-assistant-permissions/cortex-agentic-assistant-agents.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.