Anthropic proposes and open-sources the Model Context Protocol, a universal protocol for bridging data between existing apps and AI systems

Anthropic, an AI company known for its chat AI 'Claude,' has announced the Model Context Protocol (MCP), a protocol that enables seamless integration of AI systems with external data sources. MCP is being developed as open source, and detailed documentation is already available.

Introducing the Model Context Protocol\Anthropic

Introducing the Model Context Protocol (MCP)

An open standard we've been working on at Anthropic that solves a core challenge with LLM apps - connecting them to your data.

No more building custom integrations for every data source. MCP provides one protocol to connect them all: pic.twitter.com/kYsivQyPDq

— Alex Albert (@alexalbert__) November 25, 2024

AI can be extremely useful when used in combination with existing business and development tools. However, because each tool handles data differently, AI system developers need to optimize how data is used for each corresponding tool.

MCP is a protocol that acts as a bridge between AI and external data sources. If MCP becomes widespread, developers of various tools will be able to complete AI support simply by adding a function to publish data via the MCP server, and AI developers will be able to handle data from each tool simply by building a function to reference data from the MCP server. This will reduce the burden on both tool developers and AI developers.

Anthropic has released an MCP server that can be integrated with Google Drive, Slack, GitHub, Git, Postgres, Puppeteer, etc. as a reference implementation of MCP. In the video included in the following post, Claude performs tasks such as 'creating a repository on GitHub,' 'creating a branch,' and 'creating an issue' through MCP.

Here's a quick demo using the Claude desktop app, where we've configured MCP:

Watch Claude connect directly to GitHub, create a new repo, and make a PR through a simple MCP integration.

Once MCP was set up in Claude desktop, building this integration took less than an hour. pic.twitter.com/xseX89Z2PD

— Alex Albert (@alexalbert__) November 25, 2024

The MCP documentation is available at the following link:

Introduction - Model Context Protocol
https://modelcontextprotocol.io/introduction

In addition, official information related to MCP is summarized on the following page.

Model Context Protocol · GitHub
https://github.com/modelcontextprotocol