Overview: The Model Context Protocol (MCP) and MCP Servers

What Is the Model Context Protocol?

MCP is an open-source protocol, originally introduced by Anthropic, for connecting AI applications to external systems. Before MCP, connecting your AI assistant to an outside tool or data source—a calendar, a database, a file system, a search engine—typically needed its own custom integration. MCP replaces that patchwork with a single, standardized way for your AI assistant to discover and use outside capabilities.

A useful comparison is a USB-C port: just as USB-C gives your devices one standard way to connect to peripherals, MCP gives your AI assistant one standard way to connect to data sources, tools, and prompts.

How MCP Works

MCP follows a client-server model with three participants:

• MCP host — the AI application you use, such as Claude Desktop, Claude Code, or another MCP-compatible assistant. It coordinates one or more connections to outside servers, including Bluehost's.
• MCP client — a component inside your AI assistant that maintains a dedicated connection to a single MCP server.
• MCP server — a program, like Bluehost's, that exposes context and capabilities to your AI assistant. A server can run locally on your own machine, or remotely on its own infrastructure—Bluehost's runs remotely.

An MCP server can expose three kinds of building blocks, called primitives:

• Tools — actions your AI assistant can take on your behalf, such as looking up your account or purchasing a plan.
• Resources — contextual data your AI assistant can read, such as your account or hosting details.
• Prompts — reusable templates that help structure how your AI assistant interacts with a particular tool or dataset.

Together, these let your AI assistant move beyond just answering questions from its training data—it can look up your current information and take real actions on your behalf, within whatever boundaries the server defines.

What MCP Servers Are Used For

Because MCP is an open standard, a growing ecosystem of MCP servers already exists for common tools and services, including file systems, GitHub, Slack, Google Drive, and databases such as Postgres. In practice, you might use an MCP server to:

• Give your AI assistant access to your calendars, documents, or project data so it can act as a more capable, personalized assistant.
• Let a coding assistant read your design files, repositories, or issue trackers to generate or update code with real context.
• Connect a chatbot to your internal database so you and your team can ask questions in plain language.
• Automate multi-step workflows that combine several tools—for example, looking something up, transforming it, and posting the result somewhere else.

Why MCP Matters to You

As a Bluehost customer, MCP means your AI assistant can do more on your behalf:

• Less setup. Because MCP is a shared standard, you don't need a custom integration for every AI application you use—connect once, the same way, each time.
• More capability. Your AI assistant gains access to a growing ecosystem of existing servers, expanding what it can do for you without custom-built connectors.
• Real action, not just answers. Your AI assistant can be more capable and more current, and act on your behalf, rather than being limited to general knowledge.

Connecting to Bluehost's MCP Server

Bluehost runs its own MCP server at https://mcp.bluehost.com/mcp. Connect it to any MCP-compatible AI assistant, and it can look up your account and billing details and take certain actions on your behalf, right from your conversation.

To connect to the server, sign in to (or create) your Bluehost account and authorize the connector, the same way you'd authorize any other MCP connector. If you plan to purchase a Self-Managed VPS through it, make sure you already have a payment method on file—connecting the server by itself won't charge you anything. You can't add or change a payment method through the MCP itself; you'll still need to do that through your browser.

Summary

The Model Context Protocol is an open standard that lets your AI applications connect to outside tools, data, and workflows through a common set of building blocks: tools, resources, and prompts. Bluehost runs its own MCP server at https://mcp.bluehost.com/mcp, which you can connect to any MCP-compatible AI assistant to look up your account and billing details or purchase a Self-Managed VPS directly through it. At launch, this is limited to US customers paying in USD; domain purchases, Managed VPS plans, payment-method changes, and server cancellations aren't supported through the MCP yet.