MCP Integration in Claude Code
Last reviewed
Correct answer: B. Immediate availability, paid for with context that every turn of the conversation now gives up.
Explanation
The principle — Context is not storage. Whatever sits in it is paid for again on every request, so a decision to load something upfront is a decision to pay repeatedly.
Why the key is correct — Setting a server to load always means every tool from that server loads into context at session start regardless of the deferral setting. The documentation's own advice is to use this for a small number of tools that Claude needs on every turn, since each upfront tool consumes context that would otherwise be available for your conversation. You get availability without a search step, and you give up room.
Why the others are wrong — The cost is not offset by tools you would have fetched anyway, because upfront loading pays for the ones you never use as well. The setting does not depend on deferral being off; it overrides it. And the cost is not a one-time startup charge, because context is re-sent with every request.
Remember this — Always-loaded is a per-turn cost, so spend it only on tools you genuinely use per turn.
Sources — Anthropic's Claude Code MCP reference.
Sources
“Every tool from that server then loads into context at session start regardless of the ENABLE_TOOL_SEARCH setting.”
“Use this for a small number of tools that Claude needs on every turn, since each upfront tool consumes context that would otherwise be available for your conversation.”
Practise 10 questions on this topic
Take MCP Integration in Claude Code — Timed Test 1 (10 questions) — scored instantly, explanation for every question, no login.