What to know
- Anthropic has created Model Context Protocol (MCP), a tool that connects AI systems to data sources.
- Block, Apollo, Replit, and Codeium have started using MCP in their systems.
- The tool lets developers connect to data sources without writing separate code for each one.
Anthropic has launched Model Context Protocol (MCP), an open-source tool that connects AI assistants directly with data sources. The tool solves a common problem in AI development by eliminating the need for custom code for each dataset.
Alex Albert from Anthropic explains that MCP makes it easier for developers to connect their systems to data sources through a single integration, saving time and effort in the development process.
Getting LLMs to interact with external systems isn't usually that easy.
— Alex Albert (@alexalbert__) November 25, 2024
Today, every developer needs to write custom code to connect their LLM apps with data sources. It's messy, repetitive work.
MCP fixes this with a standard protocol for sharing resources, tools, and prompts.
Several tech companies have already adopted the protocol. Block and Apollo are using it, while Zed, Replit, Codeium, and Sourcegraph are adding MCP support to their platforms.
Anthropic offers ready-to-use MCP servers for Google Drive, Slack, and GitHub. Claude Enterprise users can link their chatbots to internal systems using these MCP servers.
The company wants AI systems to maintain context while working with different tools and datasets. This would create a more unified way of handling data integration.
Anthropic has made MCP open-source and welcomes developers to contribute. They can start using pre-built servers through the Claude Desktop app or follow the quickstart guide.
Source: Anthropic
Discussion