Model Context Protocol (MCP) lets AI Agents link up with external tools and resources. You can connect AI Agents from various frameworks and platforms to Automation Anywhere automations.

Using MCP, an AI Agent can:
  • Get data from third-party systems.
  • Make decisions using company data.
  • Run and discover automations from the Control Room.
    Note: The Run Automation and Discover Automation tools currently support these variables:
    • Input: String, Number, and Boolean
    • Output: String, Number, Boolean, and List

Prerequisites

Make sure you have set up the following:

  • Control Room repository permissions.
  • An MCP client: For example, Microsoft Copilot, ENTKB (Enterprise KB), or another MCP-compatible agent that can register an external MCP server and use tools. Each MCP client might support different MCP features, allowing for varying levels of integration with MCP servers. For a comprehensive list of supported MCP clients and latest features, click Feature support matrix.

Procedure

  1. Sign in to the Control Room.
  2. Navigate to Administration->Settings->Agent connections.
  3. From the MCP Inbound tools section, you can enable third-party MCP servers to use automations as tools. Select Use automations as MCP tools and then click either:
    • Use all automations in the Control Room: This options lets you access all automations without needing to set up each one separately.
    • Only use selected automations: This option allows you to choose and manage which specific automations will be available on the Agent connections page.
  4. Click Process Reasoning Engine to scan the available automations. This will help create descriptions that AI Agents can use while they are working.
  5. Click Save changes.
  6. Navigate to AI->Agent connections. From the Agent connections page, you can manage how automations and AI Agents connect with external servers and AI Agents through inbound and outbound connections.
  7. From the Agent connections page, click the Inbound tools tab. Automations added here become tools that third-party servers and AI Agents can use and run during workflows.
  8. At the top of the page, click + Add-> Inbound tool.
  9. Select files to share as tools for third-party MCP servers. In your Repository, you can pick individual files: (bots, processes, and API Tasks), or you can choose entire folders. If you choose a folder, all items inside the folder will be shared. Click the arrow to move the file from the Available list to the Selected list.
  10. Click Next to open the Tool details page. This page displays the tool's name and description, which help identify the tool when it's being used.
  11. Enter tool details:
    • Tool name (Required): Rename to something AI-friendly and make it easier for the third-party AI Agent to understand. The name must be unique across all Agent connections.
    • Descriptions (Required): Keep it clear and action-oriented (what it does, key inputs or outputs).
  12. Click Create. Your selected tool now appears in the Inbound tools table with a connection status of Available.
    Note:

    If an automation is updated or deleted, then you must update or delete the agent connection manually.