> For the complete documentation index, see [llms.txt](https://developer.emporix.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://developer.emporix.io/agentic-commerce-intelligence/agentic-intelligence/agent-library/support-agent.md).

# Support Agent

The **Support Agent** is designed to enhance internal communication and operational efficiency within your team. Through the communication bot installed in Slack or MS Teams, your team can interact with the Emporix system directly within the collaboration workspace.

When the collaboration with the Support Agent is established, the participants can ask questions about the case context, request more details or perform actions directly through the agent.

## Purpose

The agent acts as a bridge between humans and Emporix agents or connected MCP tools, empowering support teams to handle complex issues more efficiently.

## Key benefits

Enabling the Support Agent brings several key benefits for your business.

| Benefit                                   | Description                                                                                                                                                                                                                                                                          |
| ----------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
| **Seamless collaboration**                | It allows team members to inquire about orders, returns, or product updates directly in context-aware collaboration channels, fostering improved teamwork and information sharing.                                                                                                   |
| **Reduced application switching**         | By bringing Emporix system interactions into Slack or MS Teams, it eliminates the need for employees to switch between different applications, saving time and enhancing focus.                                                                                                      |
| **Increased operational efficiency**      | Automating access to real-time information within a communication tool helps streamline workflows, contributing to a more efficient operation overall.                                                                                                                               |
| **Ease of adoption and setup**            | As a prebuilt agent from the Agent Library, it offers a ready-to-use solution that integrates seamlessly into existing business workflows. The automatic one-click installation simplifies the setup process significantly, ensuring quick deployment without complex configuration. |
| **Enhanced customer-friendly operations** | By making internal information more accessible and promoting efficient team communication, it indirectly supports a more responsive and customer-friendly operation.                                                                                                                 |

## How it works

The Support Agent template creates a **generic** agent instance named **Support Agent**. When you have one Support Agent instance and no other agents connected to the communication tools, the agent installs as a bot named **Emporix Support Agent** in Slack workspace or **Emporix Agent** in Microsoft Teams team accordingly. The bot is available in the background in the communication workspace and can trigger a relevant agent (as per configuration) when needed.

When the [Complaint Agent](/agentic-commerce-intelligence/agentic-intelligence/agent-library/complaint-agent.md) or [Anti-Fraud Agent](/agentic-commerce-intelligence/agentic-intelligence/agent-library/anti-fraud-agent.md) starts collaboration, that agent creates the dedicated channel (or group chat), invites participants, posts a message, and assigns the Support Agent to the channel. The Support Agent then handles ongoing conversation in Slack or MS Teams through its attached communication tool (Slack AI Tool or MS Teams AI Tool) and the operations granted in that tool's **Allowed operations**.

### Trigger

The Support Agent can be triggered directly from your team's communication tool. Once you establish the connection and configure the agent adequately, the team can speak to the agent directly within the collaboration space. The supported triggers are **Slack** and **MS Teams**.

{% hint style="info" %}
When you select a communication trigger, attach the matching communication tool and grant the required **Allowed operations**. Without the tool, the agent cannot receive or respond to messages in the channel.
{% endhint %}

### Custom entities

When the [Complaint Agent](/agentic-commerce-intelligence/agentic-intelligence/agent-library/complaint-agent.md) or [Anti-Fraud Agent](/agentic-commerce-intelligence/agentic-intelligence/agent-library/anti-fraud-agent.md) starts collaboration for a specific case, that agent creates the dedicated communication channel and assigns the Support Agent to it. When collaboration starts, **Complaint Audit** or **Anti-Fraud Audit** also creates the Support Agent Context entity in the system.

#### Support Agent Context

Whenever collaboration is started and conversation happens on a collaboration channel, an instance of the Support Agent Context is kept in the **Custom Instances**. The Support Agent Context records the following information:

* Collaboration tool used
* Slack or MS Teams Channel ID created for the complaint case
* Conversation history
* Invited participants

## Agent configuration

To allow the agent to act directly within the company collaboration channel (Slack, MS Teams) and respond to merchants' queries about certain operations in the Emporix system (or an external one), configure the agent by connecting it to the relevant AI Tools and MCP servers.

### User Prompt

The **User Prompt** determines the general agent behavior and is fully editable. The default prompt defines the agent actions for the case when an invoice is missing and billing address needs an update:

<details>

<summary>Default User Prompt for Support Agent</summary>

````
You are a Customer Support Agent for Slack channel communication.
<objective> Help support users and teams with order issues, invoices, and business inquiries reported by customers. You are always communicating with the support users! </objective>
<execution-strategy> ALWAYS run steps sequentially - one by one <execution-strategy>
<instructions>
1. Based on the conversation history, analyze the inquiry (order/invoice/customer issue)
2. _thinking: make sure that all the steps are executing in a SEQUENTIAL way
3. Execute tools using the appropriate strategy
4. Respond professionally with clear, actionable information
CRITICAL: Analyze the attached conversation history to find relevant information or context
</instructions>
<important>
Use tools to fetch and manage order/customer data when needed
<rules>
<rule> For missing invoices: create and send new invoices </rule>
<rule> For billing updates: ALWAYS update billing first, THEN use `send-invoice` tool to send the invoice automatically </rule>
</rules>
<rule> Escalate complex issues appropriately - Maintain professional, empathetic tone </rule>
</important>
<output-format>
- Only use formatting supported by Slack.
- Do not use HTML, images, or any unsupported formatting.
- Ensure your response is always displayable and readable in Slack.
- NOT USE Markdown formatting.
</output-format>
<slack-output-requirements>
<example>
- Bold text:*your text*
- Italics: _your text_
- Code blocks: ```your text```
- Ordered list: 1. Item 1
- Lists:   • Item 1   • Item 2   • Item 3
- block quote: > your text
</example>
<rule> Be helpful, accurate, and solution-focused </rule>
</slack-output-requirements>
````

</details>

Configure how the agent works and acts by adjusting the prompt rules and including the cases you want the agent to act upon.

### Scopes

Since the agent is triggered by Slack or MS Teams directly, and not by users, there is no need for setting up any required scopes for the Support Agent.

### Capabilities and tools

The capabilities of the Support Agent are strictly connected to the MCP server tools you configure for it. Upon creation, the agent is equipped with the tools for **Customer** and **Order** entities management.

Default toolset:

| Tool                                | Capabilities                                  |
| ----------------------------------- | --------------------------------------------- |
| **Customer MCP Tools**              |                                               |
| `get-customer`                      | Retrieves customer details.                   |
| `get-legal-entity`                  | Fetches details of a specific legal entity.   |
| `get-legal-entities`                | Lists all legal entities.                     |
| `change-customer-activation-status` | Activates/deactivates a customer.             |
| `send-email-to-customer`            | Sends email to the customer email address.    |
| **Order MCP Tools**                 |                                               |
| `get-order`                         | Retrieves a specific order.                   |
| `get-customer-orders`               | Lists all orders for a customer.              |
| `update-order-billing-data`         | Updates billing information on an order.      |
| `update-order-delivery-date`        | Updates the delivery date on the order level. |
| `send-invoice`                      | Resends an invoice.                           |
| `change-return-status`              | Changes the status of a created return.       |
| `get-return`                        | Retrieves return's details.                   |

Extend or limit the tools according to your needs in the **Tools** tab. Add or remove the tools from the list.

<figure><img src="/files/eNTUY0folqzFPuugUrdN" alt="MCP Tools"><figcaption><p>MCP Tools</p></figcaption></figure>

{% hint style="info" %}
Optionally, attach an Orders [RAG Emporix Tool](/agentic-commerce-intelligence/agentic-intelligence/configuration/tools/rag/rag-emporix.md) (`RAG_EMPORIX` with entity type **Order**) for natural-language order discovery in Slack or MS Teams. Order MCP tools remain the source of truth for exact reads and updates. Retrieval is limited by available scopes — for example, tenant-wide, legal entity, customer, or legal entities available to a customer. See [Scoped retrieval for orders](/agentic-commerce-intelligence/agentic-intelligence/configuration/tools/rag/rag-emporix.md#scoped-retrieval-for-orders).
{% endhint %}

The Support Agent requires the **Slack AI Tool** or **MS Teams AI Tool** to work properly. First, create the communication tool bot in the **Agentic AI** -> **AI Tools** and choose the appropriate type.

{% hint style="info" %}
For more information on how to create a bot, see the [Communication Tools](/agentic-commerce-intelligence/agentic-intelligence/configuration/tools/comm-tools.md).
{% endhint %}

Once the relevant AI Tool is ready, establish the connection with the bot by selecting the AI Tool in the **Native Tools** list. Channel creation and related collaboration actions depend on the **Allowed operations** configured on that tool instance (for example, Create channel, Invite participants, Send message).

<figure><img src="/files/1KVhRgiDmO7Ej59sBAoW" alt="Adding native communication tool and scoping capabilities"><figcaption></figcaption></figure>

This way, the Support Agent can act directly within the Slack workspace or MS Teams team.

## Workflow

Typically, the Support Agent workflow begins when the [Complaint Agent](/agentic-commerce-intelligence/agentic-intelligence/agent-library/complaint-agent.md) or [Anti-Fraud Agent](/agentic-commerce-intelligence/agentic-intelligence/agent-library/anti-fraud-agent.md) template instances decide that human collaboration is needed. They fetch the required data, start collaboration, and assign the Support Agent to the channel. Audit persists the case context and, if collaboration was started, the Support Agent Context. Participants then communicate with the Support Agent in the created channel.

The example workflow for the Support Agent can look as follows:

{% stepper %}
{% step %}

#### Collaboration setup

The [Complaint Agent](/agentic-commerce-intelligence/agentic-intelligence/agent-library/complaint-agent.md) or [Anti-Fraud Agent](/agentic-commerce-intelligence/agentic-intelligence/agent-library/anti-fraud-agent.md) creates a dedicated Slack or MS Teams channel (or group chat), invites participants, posts a case message, and assigns the Support Agent to the channel. Audit then persists the case context and, if collaboration was started, the Support Agent Context. These actions run through the attached communication tool and its **Allowed operations**.

> For example, there is a complaint from a customer reported in relation to their recent order.
> {% endstep %}

{% step %}

#### Human interaction

The channel members write directly in the channel asking the agent for more details or requesting a specific action.

> For example, the participants ask the Support Agent about specific order details that are relevant to the complaint case.
> {% endstep %}

{% step %}

#### Action execution

The Support Agent uses its configured tools to fulfill the requests. Responses and results are shared back in the channel for full transparency.

> For example, the agent gets back with the retrieved order details and customer information. When necessary, the agent resends the invoices and updates the order details.
> {% endstep %}
> {% endstepper %}

{% hint style="info" %}
All Agents operations are tracked in the **AI Logs** view. You can check the details of the requests, jobs, and sessions for monitoring and evaluation purposes. For more information, see the [AI Logs](/agentic-commerce-intelligence/agentic-intelligence/logs.md).
{% endhint %}


---

# 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://developer.emporix.io/agentic-commerce-intelligence/agentic-intelligence/agent-library/support-agent.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.