# AI-Suggested Knowledge Articles

AI-Suggested Knowledge Articles helps agents respond quickly and accurately by recommending relevant knowledge articles during customer conversations.

Instead of manually searching, agents get real-time suggestions based on the context of the chat, making interactions more efficient and consistent.

**How It Works**

While an agent is chatting with a customer, the system analyses the conversation and automatically suggests relevant Knowledge Articles.

These suggestions appear in the Knowledge Article Panel within the chat interface, allowing agents to instantly view and use them.

### AI Prompt Configuration

This feature is powered by AI prompts that determine how knowledge articles are suggested.

* A **default system prompt** is already configured in the application
* Users can define their own **custom prompt** using the *User Instructions* field
* When both are available, **User Instructions take priority** over the default system prompt

#### How to Configure Custom Prompt

To update the prompt as per your business needs:

1. Navigate to **WhatsApp4Dynamics Application**
2. Go to AI **Prompt** entity
3. Search for **Knowledge Article Prompt**
4. Open the record
5. Update the **User Instructions** field as required
6. Save the changes

<figure><img src="/files/kl1cqOa8suPFtidOvnLi" alt=""><figcaption></figcaption></figure>

### Knowledge Article Panel

The Knowledge Article Panel is designed to support both **AI-based suggestions** and **manual search**, making it flexible and user-friendly.

#### 1. AI Suggested / Results Tab

* By default, when AI suggests articles, the tab is labelled **“AI Suggested Knowledge Articles”**
* When an agent performs a manual search, the same tab dynamically changes to **“Results”**
* This means the same tab is used to display both:

&#x20;     AI-suggested articles

&#x20;     Manually searched articles

This ensures a seamless experience without switching between multiple sections.

<figure><img src="/files/yBYYIbOFK7v6vVeFMutb" alt=""><figcaption></figcaption></figure>

### 2. Favourite Articles

* Agents can mark any article as a Favourite
* Favourite articles are stored separately for quick access
* Useful for frequently used or important articles

<figure><img src="/files/BqhAdNxKQrmrfb08LG6f" alt=""><figcaption></figcaption></figure>

### Search Knowledge Articles

Agents can also search for articles manually:

* Enter keywords in the search bar
* Relevant knowledge articles will be displayed instantly
* Useful when looking for specific information

<figure><img src="/files/juDubSw3dXXzVOCZRDpU" alt=""><figcaption></figcaption></figure>

### Actions Available

&#x20;**Copy Article Content**

* Each knowledge article has a Copy button
* Clicking it copies the content of the article
* Agents can paste it directly into the chat

&#x20;**Mark as Favourite**

* Click the Favourite button on any article
* The article will be saved under the Favourite tab

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

* Only published knowledge articles are visible in the chat UI
* Users must have the required privileges for:
* Knowledge Article entity
* Favourite Knowledge Article entity
  {% endhint %}

This feature ensures that agents always have the right information at the right time, improving both productivity and customer experience.


---

# 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://docs.inogic.com/whatsapp4dynamics/ai-integration-inowiz-copilot/features/ai-suggested-knowledge-articles.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.
