> 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-xsiam/learn-about-cortex-xsiam/agentic-ai-in-cortex-xsiam/compare-agentic-assistant-with-cortex-assistant.md).
# Compare Agentic Assistant with Cortex Assistant
Cortex XSIAM offers two distinct forms of AI-driven assistance. Agentic Assistant is an advanced, optional capability that utilizes generative AI to autonomously plan and execute complex workflows. Cortex Assistant is a basic interface for streamlined navigation and entity investigation using natural language.
The following details the differences between Agentic Assistant and Cortex Assistant.
| Features | Agentic Assistant | Cortex Assistant |
| --------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| How it operates | Uses a Large Language Model (LLM) to analyze intent and dynamically generate a plan, a unique sequence of actions executed step-by-step to resolve a specific request. | Uses natural language processing to convert user questions into XQL queries and suggest a list of static, predefined responses (for example, "Run Playbook," "Scan Host"). |
| Scope of operation | Complex, ad-hoc scenarios. Agents function as virtual personas (for example, Threat Intel, IT) that can autonomously determine the necessary steps to achieve a broad objective. | Routine tasks such as single-entity investigations (host, hash, user) and navigation shortcuts. |
| Customization | Anyone with the relevant permissions can build custom agents with specific instructions, personas, and restricted sets of actions. Scripts and commands can be registered as new actions for agents to utilize. | Functionality is limited to out-of-the-box capabilities provided by the platform. You cannot modify Cortex Assistant's behavior. |
| Execution logic | Agents validate their own plans, clarify ambiguous prompts, and execute multiple steps in sequence or parallel based on the context of the investigation. | Relies on traditional rule-based automation. Actions are discrete and require manual selection from a recommended list. |
| Infrastructure | Leverages dedicated Google Cloud Platform (GCP) infrastructure for GenAI processing. | Processes queries within the standard tenant infrastructure. |
| Availability | Disabled by default. It requires enablement by an Administrator via Settings → **Configurations** → **General** → **Server Settings** → **Agentic Assistant** and is currently restricted to tenants in specific regions. For more information, see [Cortex Agentic Assistant](/cortex-xsiam/learn-about-cortex-xsiam/agentic-ai-in-cortex-xsiam.md). | Available by default to all tenants not using Cortex Agentic Assistant. |
| Access Control (RBAC) | Administrators use a dedicated CORTEX AGENTIC ASSISTANT permission category to configure specific permissions for:
- Interacting with agents using the chat interface.
- Managing agents/actions: Viewing, creating, or editing custom agents and registering new actions in the Agentic Assistant Hub.
| Permissions are determined by standard Cortex XSIAM user roles (for example, View/Edit access to specific modules). |
| Auditing | All agent activities are logged in a specific dataset (**`agentix_agents_actions`**) queryable via XQL. You can also view the specific plan generated by the AI within the chat interface to understand the logic behind an answer. | Actions taken are logged as standard system activities. |
---
# 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-xsiam/learn-about-cortex-xsiam/agentic-ai-in-cortex-xsiam/compare-agentic-assistant-with-cortex-assistant.md?ask=&goal=
```
`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.