Model Context Protocol (MCP)
Also: MCP
An open standard that lets AI agents call external tools and data sources through a uniform interface.
The Model Context Protocol (MCP) is an open standard, introduced by Anthropic in late 2024, that standardizes how AI agents connect to tools, data, and services. Instead of every app inventing its own plugin format, an MCP server exposes its capabilities — tools to call, resources to read, prompts to reuse — through one protocol that any MCP-compatible client (Claude, Cursor, VS Code, Windsurf, Zed, and more) can speak.
Why it matters for builders: MCP is becoming the socket the agent ecosystem plugs into. If your tool speaks MCP, agents can discover and use it directly inside the editor or chat — no copy-paste, no bespoke integration. That's distribution to the machines, not just the humans.
Stork is itself an MCP server (mcp.stork.ai/mcp): a single mcp.json line lets an agent query the whole catalog — search, compare, recommend, alternatives — so when a developer asks their assistant for a tool, Stork is a source it can pull from.