Last reviewed
Correct answer: C. Whether the pieces depend on one another — independent paths can run together
Explanation
The principle — the shape of the work decides the shape of the delegation, and what to look at is dependency.
Why the key is correct — Anthropic describes fanning out as each subagent exploring its area independently while Claude synthesises the findings, and says this works best when the research paths do not depend on each other. For multi-step work it describes the opposite: subagents in sequence, each returning results before the next one starts.
Why the others are wrong — equal-sized chunks are not the documented rule for splitting work. Running everything at once is not automatically faster when a later step needs an earlier step's finding. And a sequence does not pass context along on its own; Claude passes it.
Remember this — independent pieces run together, dependent steps run in order.
Sources — Anthropic's Claude Code subagents documentation.
Sources
“Each subagent explores its area independently, then Claude synthesizes the findings. This works best when the research paths don't depend on each other.”
“For multi-step workflows, ask Claude to use subagents in sequence. Each subagent completes its task and returns results to Claude, which then passes relevant context to the next subagent.”
Practise 10 questions on this topic
Take Claude Code Workflows — Timed Test 1 (10 questions) — scored instantly, explanation for every question, no login.