> 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/configure-cortex-xsiam/configure-the-cortex-agentic-assistant-1/agents-hub/manage-knowledge-sources-preview.md).

# Bring your own knowledge

Use knowledge sources to give Cortex XSIAM Agentic Assistant agents trusted business context.

The Knowledge center in the Agentic Assistant Hub enables your agents to act as personalized extensions of your team by grounding them in your unique business context. By connecting your own documentation and procedures to the Cortex Agentic Assistant, you turn AI agents into experts on your environment rather than relying on general information..

### Grounding your agents

The Agentic Assistant uses two types of knowledge to provide precise, context-aware results:

* Organizational-specific (custom) knowledge: Documentation you provide to the platform. You can upload files (such as SOPs, policy docs, and analyst notes) or ingest specific web pages to teach the agent about your internal environment and procedures.
* Cortex (system) knowledge: Built-in product expertise provided by Palo Alto Networks. This is natively part of the platform and gives the agent an intrinsic understanding of Cortex security entities, data schemas, and UI workflows.\
  System knowledge is built-in and cannot be manually added or modified by users.

### How agents use your knowledge

When you ask the agent to perform a task, it checks your connected documentation to understand the why and how of your organization's specific processes.

* Personalized investigations: The agent uses your documents to guide its work. For example, if your SOP contains guidance on how to handle a specific type of lateral movement, the agent factors that context into its investigation plan.
* Actionable context: Beyond just answering questions, your knowledge informs how the agent uses its technical tools. It identifies the most relevant skills and sequences them according to the logic found in your documentation.
* Suggested prompts: After providing a response, the agent often suggests follow-up questions to help you continue the investigation based on the context it retrieved.

### Transparency and grounding

The platform provides visibility into how knowledge informs an agent's response.

* Searching knowledge sources: During a chat session, you can see a status indicator showing that the agent is actively searching your knowledge base to inform its logic.
* Citations and references: The agent includes clickable clips within its responses to show which document supported a specific point. An expandable list of References is provided at the bottom of the response, identifying all custom and system sources used.
* Conflict signaling: If your custom documentation provides guidance that differs from built-in system knowledge, the agent prioritizes your organizational knowledge by default and signals that conflicting guidance was detected.
* Insufficient grounding: If the agent cannot find enough trusted context to answer safely, it will tell you that grounding is insufficient rather than guessing.

{% hint style="info" %}
Note:

To add and manage knowledge sources, you must have view/edit permissions. For more information, see [Agentic Assistant role-based access control](https://cortex-docs.paloaltonetworks.com/cortex-xsiam/configure-cortex-xsiam/configure-the-cortex-agentic-assistant-1/agentic-assistant-role-based-access-control.md).
{% endhint %}

### Add your organizational knowledge

You can add a knowledge source by uploading a file or linking to external documents.

#### Source guidelines

* Supported file formats: MD, JSON, JSONL, CSV, DOC, or DOCX.
* File size: Up to 5MB per file.
* Ingestion: Supports single URL page ingestion.
* Capacity: If the maximum source limit is reached, you must remove existing sources before adding new ones.

#### Add a source

1. In the **Agentic Assistant Hub** → **Knowledge Center** tab, click **+ New Source**.
2. Select a source type: **File** or **Link**.

**Upload a file**

1. Drag and drop or browse to upload a supported text file.
2. Set the **Source name** (defaults to the file name).
3. In the **Shared agents** dropdown, select one or more agents (or All agents to include any future agents automatically).
4. Click **Add source**.

**Add a link**

1. In the **Data source** dropdown, select an internal knowledge source, such as Atlassian Confluence Cloud or Google Drive.
2. Select the **Instance** (the data source instance). If the instance is not configured for this data source, the source is greyed out and you are prompted to add a new instance.
   * Confluence: The URL of the Confluence page to retrieve content from.\
     Supported formats include `https://<site>.atlassian.net/wiki/spaces/<space>/pages/<pageId>/<title>` and `https://<site>.atlassian.net/wiki/pages/viewpage.action?pageId=<pageId>`.
   * Google Drive: The Google Drive file URL. Maximum file size is 5 MB. The file must be shared with the logged-in user's email.
   * O365 File Management (OneDrive/SharePoint/Teams): The SharePoint/OneDrive sharing URL of the file. Maximum file size is 5 MB.
3. The **Source name** is by default the knowledge source you are linking to. You can change it.
4. In the **Shared agents** dropdown, select one or more (or all) agents to apply the knowledge source to. If you select all, any subsequently created agents will automatically be included.
5. Toggle **Auto sync** to automatically sync with the integration instance once every 24 hours.
6. Click **Add source**.

### Manage sources

In the **Knowledge Center** table, you can monitor your storage capacity and track source status (Processing, Connected, or Error). Right-click any source in the table to:

* **Preview**: View a text/markdown-style preview of the content.
* **Enable/disable**: Toggle a source's availability. Disabled sources remain in the list but are ignored by the agent during investigations.
* **Delete**: Permanently remove a source. This requires Knowledge Admin permissions.


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://cortex-docs.paloaltonetworks.com/cortex-xsiam/configure-cortex-xsiam/configure-the-cortex-agentic-assistant-1/agents-hub/manage-knowledge-sources-preview.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.
