Plan-then-act with explicit sub-task scaffolding

Two-turn pattern: first turn produces a numbered plan; second turn executes each sub-task and returns structured results.

The prompt

Copy this verbatim. Replace the {{ … }} placeholders with your values.

<!-- TURN 1 (plan) -->
<instructions>
You will receive a complex task. Decompose it into 3–7 sub-tasks. Return JSON inside <plan> tags:

{
  "summary": "one-sentence restatement of the goal",
  "subtasks": [
    { "id": "S1", "title": "string", "depends_on": [], "deliverable": "string" }
  ]
}

Rules:
- Each subtask must have a single, concrete deliverable (an artifact, a list, a number).
- depends_on references other subtask IDs that must complete first; empty list means parallelizable.
- Do not execute any subtask in this turn. Plan only.
</instructions>

<task>{{ task_description }}</task>

Return inside <plan> tags.

<!-- TURN 2 (execute) — re-injects the plan as context -->
<instructions>
The plan from turn 1 is in <plan>. Execute every subtask whose dependencies are satisfied
by what's already in <results>. For each, append a new <result> tag:

<result subtask_id="S1">
  ... your output for this subtask ...
</result>

After processing all currently-executable subtasks, output <status>done</status> if every
subtask in the plan now has a <result>, or <status>continue</status> if more turns are needed.
</instructions>

<plan>{{ plan_from_turn_1 }}</plan>

<results>
{{ results_from_prior_turns }}
</results>

Sample input

Task: Produce a Q3 board update for our SaaS startup covering financials, product, hiring, and risks.

Expected output

<plan>
{
  "summary": "Produce a Q3 board update with financials, product, hiring, and risks sections.",
  "subtasks": [
    {"id": "S1", "title": "Financials section", "depends_on": [], "deliverable": "MRR, ARR, burn, runway with QoQ deltas"},
    {"id": "S2", "title": "Product highlights", "depends_on": [], "deliverable": "3 shipped features with adoption metrics"},
    {"id": "S3", "title": "Hiring update", "depends_on": [], "deliverable": "headcount delta + open roles"},
    {"id": "S4", "title": "Risks & asks", "depends_on": ["S1","S2","S3"], "deliverable": "3 risks + 1 specific board ask"},
    {"id": "S5", "title": "Executive summary", "depends_on": ["S1","S2","S3","S4"], "deliverable": "150-word TL;DR"}
  ]
}
</plan>

[Turn 2 then executes S1, S2, S3 in parallel; turn 3 executes S4; turn 4 executes S5.]

Notes & tuning tips

• The dependency graph lets you parallelize: any subtask whose `depends_on` is empty (or all-completed) can run concurrently.
• Force structured plans — free-prose plans don't drive execution reliably.
• Hard-cap turns (max 4–5) and treat continue with no progress as a stop signal.
• For agent-style execution where each subtask may need tools, combine with the API's tool use feature.

What this example uses

Tags: <instructions> <format>

Patterns: agentic tool use structured output

More like this