decoupled.io
← Glossary
Definition

Model Context Protocol (MCP)

What is Model Context Protocol (MCP)?

Model Context Protocol (MCP) is an open standard that defines how AI models connect to and interact with external tools, data sources, and services. Created by Anthropic, MCP provides a consistent interface for AI assistants to perform actions beyond text generation — such as reading files, querying databases, managing content, or calling APIs.

How MCP Works

MCP follows a client-server model. An MCP server exposes a set of tools — discrete actions that an AI model can invoke. Each tool has a name, a description of what it does, and a schema defining its inputs and outputs. An MCP client (typically an AI assistant or agent) discovers available tools, understands their capabilities, and calls them as needed to complete tasks.

The key insight behind MCP is standardization. Before MCP, every AI integration required custom code — bespoke API wrappers, prompt engineering to handle tool outputs, and per-service authentication logic. MCP provides a single protocol that any tool provider can implement, and any AI client can consume.

A typical MCP interaction looks like this:

  1. The AI client connects to an MCP server and retrieves the list of available tools.
  2. When a user asks the AI to perform a task, the model selects the appropriate tool and constructs the required input.
  3. The MCP client invokes the tool on the server.
  4. The server executes the action (e.g., creating a content item, querying a database) and returns the result.
  5. The AI model incorporates the result into its response.

Why MCP Matters

MCP is significant because it turns AI assistants from passive text generators into active agents that can interact with real systems. Instead of describing how to do something, an AI model with MCP access can actually do it — create a blog post, update a product listing, or query analytics data.

For developers, MCP reduces integration effort. Rather than building custom plugins for every AI platform, you implement the MCP server specification once. Any MCP-compatible AI client can then use your tools.

MCP and Decoupled.io

Decoupled.io provides an MCP server with 25+ tools that give AI models direct access to Drupal's content management capabilities. Through these tools, an AI assistant can:

  • Create, read, update, and delete content
  • Manage taxonomies and content relationships
  • Query content by type, status, or field values
  • Handle media and file uploads
  • Manage users and permissions

This means content teams can use AI assistants like Claude to manage their CMS through natural language conversation, while developers can build agentic workflows that automate content operations.

To learn more about setting up and using MCP with Decoupled.io, see the MCP documentation.