MCP Server Overview

Updated
  • Updated on Dec 12, 2025
  • Published on Dec 9, 2025
  • 2 minute(s) read
Prev Next

Overview

The MCP Server feature in Decisions allows the platform to act as a Model Context Protocol (MCP) server. MCP is an open protocol that enables Large Language Models (LLMs) to interact with external tools and systems by exposing structured “tools” that an LLM can call. By supporting MCP, Decisions empowers AI models - such as those used through Visual Studio Code or other MCP-enabled clients to call AI Tool Flows directly, supply inputs, and receive processed outputs in real time.

Note
MCP Server functionality requires Decisions Version 9.18 or later.

With MCP support, Decisions users can expose AI Tool Flows as callable tools, enabling seamless integration between Decisions and AI agents that follow the MCP specification, such as Visual Studio Code MCP extensions.

What is MCP?

Model Context Protocol (MCP) is an open protocol that defines how AI systems communicate with external tools or data sources. It standardizes how tools are described, invoked, and authenticated, enabling AI applications to access enterprise systems safely and consistently. In the Decisions platform, MCP is used specifically to expose AI Tool Flows as callable MCP tools. Authentication is secured using Decisions Named Sessions.

How MCP Works in Decisions

When users create an MCP Server in Decisions, they associate one or more AI Tool Flows with that server. These flows act as tools exposed to external LLMs. MCP clients can then:

  • Identify available tools (AI Tool Flows)
  • Send input parameters defined in the flow
  • Execute the flow remotely and receive the output
AI Tool Flow Requires An Output

For the AI Tool flow to work with as a tool for an MCP server, the flow needs to have an output. This can be as simple as a "Complete" string or as complex as data for the application calling the MCP server to recieve.




To enable this integration, Decisions generates a dedicated MCP Server URL based on the environment’s base URL and the server’s configured element slug. This URL is automatically displayed in the MCP Server settings.

Once configured, external AI tools can connect to Decisions using this MCP Server URL along with a Named Session ID for authentication.

Creating an MCP Server in Decisions

Follow the steps below to configure an MCP Server and enable communication with external MCP clients.

1. Create an AI Tool Flow

Before configuring an MCP Server, you must create at least one AI Tool Flow that will be exposed as an MCP tool.

  1. Create or open a Designer Project.
  2. Select Create Flow. A pop-up window will appear with options for flow types.
  3. Navigate to Advanced > AI Tool Flow.
  4. Name the flow and select Create.

This creates an AI Tool Flow whose behavior is designed to work with MCP. 

2. Create the MCP Server

After the AI Tool Flow is ready, create the MCP Server that will expose it.

  1. Navigate within the project to Public > Integrations > MCP Servers.
  2. Select Add MCP Server.
  3. In the popup window, configure the following:
    • Name
    • Description (optional)
    • Element Slug – determines the unique MCP Server URL.
    • Tools – select the AI Tool Flow(s) created in Step 1.

After saving, the MCP Server URL is automatically displayed. It follows the format:

http://{portalBaseUrl}/mcp/{elementSlug}

Example:

http://localhost:80/Primary/mcp/myserver

This URL is required for MCP configuration in external clients.

3. Create a Named Session for Authentication

MCP clients require a Named Session ID to authenticate with the Decisions environment.

  1. Create a Named Session.
  2. Right-click the Named Session and select Get Named Session Value.
  3. Copy the generated Session ID.

This Named Session ID, along with the MCP Server URL, is used in the mcp.json file (or equivalent configuration) for connecting MCP-enabled tools to the Decisions platform.