# What is MCP?

#### Definition and Background of MCP <a href="#definition-and-background-of-mcp" id="definition-and-background-of-mcp"></a>

MCP, or Model Context Protocol, is an open standard designed to connect AI models and AI assistants with external tools and data sources. MCP was launched by Anthropic in November 2024 with the goal of enabling cutting-edge models to generate more accurate and relevant responses, addressing the limitations of AI models. MCP resolves the challenge in traditional AI integration, where custom code must be written for each data source, allowing AI systems to dynamically access external resources.

#### How MCP Works <a href="#how-mcp-works" id="how-mcp-works"></a>

MCP employs a client-server architecture, where AI models act as clients connecting to MCP servers. These servers serve as bridges, interfacing with various external systems such as content repositories, business tools, and development environments. MCP enables AI models to dynamically discover and utilize tools without requiring hard-coded knowledge for each integration. This is akin to the universal connectivity of a USB-C port, allowing AI models to access external data as effortlessly as devices connect to one another.Specifically, the core components of MCP include:

* **MCP Clients/Hosts**: AI models such as Claude or GPT that initiate requests.
* **MCP Servers**: Programs that provide access to files, databases, APIs, and other functionalities.
* **Communication Method**: Supports bidirectional communication, enabling AI not only to retrieve data but also to trigger actions in external systems, such as updating documents or sending emails.


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.arrowmcp.com/about/what-is-mcp.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.