Bryan Burnett

Updated 2026-07-28

The Model Context Protocol (MCP) is an open standard and open-source framework introduced by Anthropic in November 2024 to standardize the way artificial intelligence (AI) systems, particularly large language models (LLMs), integrate with and share data from external tools, systems, and data sources. It provides a standardized interface for reading files, executing functions, and handling contextual prompts, effectively acting as a universal bridge between AI models and the external world.

Background

MCP was announced by Anthropic on November 25, 2024, as an open standard for connecting AI assistants to data systems such as content repositories, business management tools, and development environments. The protocol was created at Anthropic by engineers David Soria Parra and Justin Spahr-Summers. It was designed to address the challenge of information silos and legacy systems, replacing the need for custom connectors for each data source or tool—a problem Anthropic described as an "N×M" data integration problem. MCP re-uses the message-flow ideas of the Language Server Protocol (LSP).

In December 2025, Anthropic donated MCP to the Agentic AI Foundation (AAIF), a directed fund under the Linux Foundation co-founded by Anthropic, Block, and OpenAI, with support from other companies.

Architecture

MCP defines a clear client-server architecture with three primary components:

MCP Host: An AI application or environment (e.g., an AI-powered IDE or conversational AI) that contains the LLM and serves as the user's interaction point. The host uses the LLM to process requests that may require external data or tools.
MCP Client: Located within the MCP host, the client facilitates communication between the LLM and MCP servers. It translates the LLM's requests for the protocol and converts server replies for the LLM. It also discovers and connects to available MCP servers.
MCP Server: An external service that provides context, data, or capabilities to the LLM. Servers connect to external systems like databases, web services, and APIs, translating their responses into a format the LLM can understand.

Communication between client and server uses JSON-RPC 2.0 as the transport protocol, with two primary transport methods:

Standard input/output (stdio): Used for local resources, offering fast, synchronous message transmission.
Server-sent events (SSE): Preferred for remote resources, enabling efficient, real-time data streaming.

How It Works

Each MCP server provides one or more tools (e.g., database access, calculators, code repository access) or resources (e.g., FAQ documents). The MCP client queries its server for a list of available tools and resources; the server responds with natural-language descriptions of each tool's capabilities and the expected format for calling it. This information is provided to the LLM. If the LLM requires a tool's service, the MCP host instructs the relevant MCP client to call the tool. The MCP server performs the action and returns results, which the MCP host injects into the LLM conversation.

Features

Standardized framework for integrating AI systems with external data sources and tools.
SDKs released in multiple programming languages, including Python, TypeScript, C#, Java, Kotlin, Go, PHP, Perl, Ruby, Rust, and Swift.
MCP Apps: An official extension to the protocol (built on mcp-ui) that standardizes the delivery of interactive user interfaces—such as dashboards, forms, and data visualizations—from MCP servers to host applications.
Enables applications such as querying structured databases with plain language (natural language data access).

Adoption

Following its announcement, MCP was adopted by major AI providers and technology companies:

OpenAI officially adopted MCP in March 2025 and added support for MCP in ChatGPT apps in September 2025.
Google DeepMind embraced the standard in April 2025.
Microsoft integrated MCP with Semantic Kernel and Azure OpenAI.
Cloudflare enabled deployment of MCP servers on its platform.
Salesforce integrated MCP into its Headless 360 platform.
Development tools including Zed, Replit, Codeium, Sourcegraph, Cursor, and Visual Studio Code adopted MCP to grant AI coding assistants real-time access to project context.
In April 2026, the AAIF held the MCP Dev Summit North America in New York City, drawing approximately 1,200 attendees.

Security Considerations

In April 2025, security researchers published analyses identifying outstanding security issues with MCP, including prompt injection and poisoned tools that could allow data exfiltration through connected tools. Key security principles include user consent and control, data privacy, tool safety, secure output handling, supply chain security, and monitoring/auditing.

Reception

MCP has been described as "USB-C for AI" (Ars Technica) for its role in standardizing AI-to-data connections. It has been likened to OpenAPI, a similar specification for describing APIs. The protocol addresses a growing demand for AI agents that are contextually aware and capable of pulling from diverse sources.