Last reviewed
Correct answer: C. The subagent's own description of the kind of task it handles
Explanation
The principle — a subagent is chosen by what it says it is for, so the description is the field that matters.
Why the key is correct — Anthropic states it twice over. Claude automatically delegates tasks based on the task description in your request, and a task that matches a subagent's description goes to that subagent. Yours says what the job is; the worker's says what it takes on.
Why the others are wrong — no naming is needed for delegation to happen. A worker's name is not what gets compared, so a vague one costs nothing and a vivid one buys nothing. And an @-mention controls which subagent Claude invokes, not what prompt it receives.
Remember this — write the description for the matcher; that is the field it reads.
Sources — Anthropic's Claude Code subagents documentation.
Sources
“Claude automatically delegates tasks based on the task description in your request”
“When Claude encounters a task that matches a subagent's description, it delegates to that subagent, which works independently and returns results.”
“The @-mention controls which subagent Claude invokes, not what prompt it receives.”
Practise 10 questions on this topic
Take Claude Code Workflows — Timed Test 1 (10 questions) — scored instantly, explanation for every question, no login.