one year on
Anthropic open-sources Model Context Protocol, a universal standard for connecting AI assistants to tools and data
The protocol aims to replace fragmented integrations with a single, open standard, letting AI models access data from GitHub, Slack, Postgres, and more through pre-built servers.
Nov 25, 2024 — Anthropic today open-sourced the Model Context Protocol (MCP), a proposed standard for connecting AI assistants to the systems where data lives, including content repositories, business tools, and development environments. The protocol aims to help frontier models produce better, more relevant responses by giving them access to data that is currently trapped behind information silos and legacy systems.
MCP provides a universal, open standard that replaces fragmented integrations with a single protocol. Developers can expose data through MCP servers or build AI applications (MCP clients) that connect to those servers. Anthropic released the specification, SDKs, local MCP server support in Claude Desktop apps, and an open-source repository of pre-built MCP servers for enterprise systems like Google Drive, Slack, GitHub, Git, Postgres, and Puppeteer.
Early adopters include Block and Apollo, while development tools companies Replit, Codeium, and Sourcegraph are working with MCP to enhance their platforms. Block CTO Dhanji R. Prasanna said open technologies like MCP are bridges that connect AI to real-world applications. TechCrunch notes the move comes as OpenAI brings similar data-connecting features to ChatGPT through its ‘Work with Apps’ capability, but pursues implementations with close partners rather than open sourcing. It remains to be seen whether MCP will gain traction among rivals, the report adds.
The protocol was created at Anthropic by David Soria Parra and Justin Spahr-Summers.
The record
Demonstrated Claude using MCP to connect directly to GitHub, create a repo, and make a pull request, claiming the integration took under an hour after setting up MCP in the desktop app.
Block's CTO called MCP an open technology that bridges AI to real-world applications, ensuring accessible and transparent innovation.
One year later — open only if you can handle spoilers
Within a year, MCP became the de facto standard for agent-tooling, adopted by major platforms like Microsoft and Google, and was integrated into virtually every major AI development framework. It is now considered the sleeper release of 2024, quietly solving the connectivity problem that had plagued agentic AI.