> For the complete documentation index, see [llms.txt](https://kb.bevica.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://kb.bevica.com/copilot-ai/copilot-and-ai/copilot-agents.md).

# Copilot Agents

Learn how Copilot agents automate defined tasks in Business Central.

Copilot agents extend AI assistance into process automation. They perform defined tasks with minimal human input. Their activity remains transparent and reviewable.

> <mark style="color:$info;">**Copilot helps you do the work, an agent can do defined work for you.**</mark>

### Standard agents in Business Central

#### [Sales Order Agent](https://learn.microsoft.com/en-us/dynamics365/business-central/sales-order-agent?toc=/dynamics365/business-central/toc.json)

The Sales Order Agent monitors customer sales requests received by email. It identifies the customer, checks availability, and prepares or refines a sales quote for review. After customer confirmation, it converts the quote into a sales order.

#### [Payables Agent](https://learn.microsoft.com/en-us/dynamics365/business-central/payables-agent?toc=/dynamics365/business-central/toc.json)

The Payables Agent monitors a designated company mailbox for PDF vendor invoices. It extracts invoice details and matches them to vendors, purchase orders, and receipts. For unmatched lines, it proposes purchase order matches, account classifications, and posting accounts. It creates draft purchase documents and routes exceptions for review.

#### [Expense Agent](https://learn.microsoft.com/en-us/dynamics365/business-central/expense-management/expense-management-overview)

The Expense Agent supports employee expense management. It extracts merchant, amount, date, and category details from a receipt. It also checks expenses against company policy before submission. As an employee, you can access it through its web app, email, or Copilot chat without a Business Central licence.

### Custom agents for Business Central

You can build agents that connect to Business Central with Microsoft Copilot Studio. Two connection methods are available.

#### Business Central connector

The Dynamics 365 Business Central connector provides predefined actions for low-code integrations. Add these actions as agent tools to find, create, or update Business Central records.

#### Business Central MCP server

The Business Central Model Context Protocol (MCP) server exposes tools generated from configured API pages and API queries. MCP configurations define the APIs and permitted operations available to an agent.

The built-in default MCP configuration provides read-only access to eligible API pages and API queries. To create, update, or delete data, or invoke API actions, use an MCP configuration that allows those operations.

#### Use both connection methods

An agent can use the connector and MCP server together. This combines dynamically discovered MCP tools with explicitly configured connector actions.

### What connected agents can do

The tools, APIs, and permissions available to an agent define what it can do. Connected agents can:

* Present Business Central customers, orders, invoices, and inventory.
* Find, create, and update records.
* Create sales orders
* Check credit
* Post payments
* Trigger approvals or flows.
* Call custom APIs exposed by AL extensions.
* Apply server-side Business Central logic, such as pricing, discounts, and validation rules.

{% hint style="warning" %}
Agents use the signed-in user's Business Central permissions. Give each agent only the data access and operations it needs. Keep appropriate review and approval controls.
{% endhint %}

### Microsoft guidance

{% embed url="<https://learn.microsoft.com/en-us/dynamics365/business-central/dev-itpro/ai/create-agent-in-copilot-studio>" %}

{% embed url="<https://learn.microsoft.com/en-us/dynamics365/business-central/dev-itpro/ai/mcp-overview>" %}


---

# 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://kb.bevica.com/copilot-ai/copilot-and-ai/copilot-agents.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.