# Knowledge base

In addition to individual customer questions, you can provide your AI Agent with a comprehensive knowledge base. This can contain content from various documents, for example FAQs, product information, or process descriptions. The AI phone assistant uses these documents to recognize connections and derive answers independently. The content of the knowledge base can be updated at any time or replaced with new documents.

<figure><img src="/files/389036d2b8b328f6a3be4ae167190eb2ff0a778d" alt=""><figcaption></figcaption></figure>

### Preparing documents optimally for the knowledge base

So that your AI Agent can process the uploaded documents in the best possible way, you should pay attention to the following points when preparing them.

#### **Structure & format**

* Use a clear structure with meaningful headings
* Use plain text instead of scanned documents
* Avoid tables where possible or convert them into plain text
* Images and graphics are not evaluated by the AI Agent

#### **Content optimization**

* For extensive content, divide it into several topic-specific documents
* Automatically generated document titles should not be manually changed
* Avoid duplicates with customer questions already stored

### Automatically captured knowledge

When creating an AI Agent, relevant information from the company website stored during the sign-up process can be automatically transferred. This content serves as the starting point for the agent's context and customer questions.

### Expand and update the knowledge base

The knowledge base of your AI Agent can be expanded or adjusted at any time. New documents, updated FAQs, or changed product information are automatically taken into account in future conversations and continuously improve answer quality.

If the AI Agent's answers are incomplete or not appropriate, you can specifically refine the underlying content. Clearly separated topics and unambiguous wording in the documents ensure that the phone assistant can reliably assign answers and output them correctly.

### Knowledge base vs. playbooks – What is the difference?

The knowledge base and playbooks fulfill different functions in the AI Agent.

{% columns %}
{% column %}
**Knowledge base:**\
Determines, **what** the AI Agent can say.

That is, which information, facts, and answers are available to it.
{% endcolumn %}

{% column %}
**Playbooks:**\
Define, **how** the AI Agent proceeds.

That is, which steps are carried out in a conversation, for example capturing callback data or forwarding a request to an employee.
{% endcolumn %}
{% endcolumns %}

Both areas interact and complement each other: the knowledge base provides the content, while playbooks control the structure and flow of the conversations.

{% embed url="<https://www.loom.com/share/2c635aa8fb2f4d6986b70f945eeaaddd>" %}

{% hint style="info" %}
**Note:** The AI Frontdesk is now called AI Agents.
{% endhint %}


---

# Agent Instructions: 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:

```
GET https://help.sipgate.de/ai-agents/en/behavior/wissensdatenbank.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.