MCP I Use

Model Context Protocol (MCP) is an open standard for connecting AI assistants to external tools and data sources.

MCP defines a clear set of standard capabilities for servers (external tools) to expose — tools, resources, prompts, sampling, elicitation, and roots — but most clients (AI assistants) only support a subset of these features.

In the software engineering world, the clients that matter most are IDEs and CLI tools attached to LLMs, and the picture is messy. VS Code with GitHub Copilot supports a different set of MCP features than VS Code with CLINE, which again differs from JetBrains or terminal-based setups.

MCP I Use collects compatibility tables for these environments, so it’s easy to see which MCP features are available before relying on them in MCP servers or workflows.