What is Model Context Protocol (MCP) and what does it enable?	MCP is an open standard that lets cloud code connect to external tools and data sources, allowing it to automatically understand when to use those tools to better answer queries.
How does agentic AI differ from traditional AI in terms of tools?	Agentic AI like cloud code uses tools to perform actions and complete tasks, whereas traditional AI typically just returns text output directly.
What are the two main types of MCP servers?	HTTP servers for remote services hosted by providers and connecting over the network, and STDIO servers for local processes running on your machine.
What command is used to add MCP servers in cloud code?	The `cloud MCP add` command is used to add MCP servers.
What are the three scopes for MCP servers?	Local (current project only), user (across all projects), and project (using a .mcp.json file checked into version control).
Why should you monitor and disable unused MCP servers?	MCP servers add tool definitions to your context window even when not in use, so disabling unused servers preserves available context.
What is more context-efficient than using MCP servers?	Using CLI equivalents (like GH for GitHub or AWS for AWS) is more context-efficient because they don't add persistent tool definitions.
What is a skill and how does it differ from MCP servers?	A skill has a name and description loaded into context that Claude loads on-demand when needed, making it more efficient than MCP servers which are always present.
What happens when MCP tools exceed 10% of your context window?	Claude code automatically switches to tool search mode, which discovers the right tools on demand.
How can your team automatically get the same MCP servers configured?	By scoping servers to project level using a .mcp.json file that you check into version control, so anyone on the codebase gets the same servers automatically.