Anthropic open-sources MCP as a universal AI-to-data connection standard
Introducing the Model Context Protocol
Anthropic announces the Model Context Protocol for two-way links between AI tools and data sources via servers and clients.
Based on
Introducing the Model Context Protocol
Anthropic frames MCP as an open standard so AI systems reach data behind silos and legacy systems without a custom integration per source. Developers expose data via MCP servers or build MCP clients that connect to those servers. The launch bundles the specification and SDKs, local MCP server support in Claude Desktop, and a repository of servers, including pre-built ones for Google Drive, Slack, GitHub, Git, Postgres, and Puppeteer.
For AI and platform teams, one protocol lowers the cost of wiring assistants into internal tools and datasets while keeping the connection pattern consistent.
A shared client–server contract for context is interoperability of meaning at the tool boundary: agents read from the same protocol surface rather than inventing private adapters to each system.
Put this to work on CoreModels
Related connectors and recipes
See all connectors →Take the next step
Try CoreModels, talk with our team, or explore more resources.