Model Context Protocol (MCP)

Before MCP, every AI assistant needed its own custom integration for every tool: one connector for GitHub in one app, a different one in the next. MCP replaces that grid with a single protocol. A tool is wrapped once as an MCP server, and any assistant that speaks MCP can use it. The project's own docs compare it to a USB-C port for AI applications.

MCP has three roles. The host is the AI application the person uses, such as a chat app or a code editor. Inside it, an MCP client holds one connection to each MCP server, and the server exposes capabilities. Messages are JSON-RPC 2.0. The specification defines two standard transports: stdio, for a server running as a local process, and Streamable HTTP, for a remote server.

A server can offer three kinds of capability:

• Tools: actions the model can call, such as "create an issue" or "run a query".
• Resources: data the application can read into context, such as files or records.
• Prompts: reusable templates the user can pick.

Clients can offer features back to servers, such as sampling (asking the host's model to generate text) and elicitation (asking the user a question).

Anthropic published MCP as an open specification in November 2024. In December 2025 it donated the protocol to the Agentic AI Foundation, a directed fund under the Linux Foundation co-founded by Anthropic, Block and OpenAI. The specification is versioned by date (for example 2025-11-25), and clients and servers agree on a version when they connect.

Vibely uses MCP in three directions. The Vibely MCP server lets Claude, ChatGPT, Cursor, Claude Code, VS Code and Codex build, edit, review and publish your Vibely apps, with 52 tools behind OAuth sign-in. Custom MCP connectors go the other way: paste a remote MCP server's URL and the Vibely agent can use your own tools and data while it builds. And agent integrations turn a published Vibely web app into an MCP server of its own, so your users can work with your app from their AI assistant.