You will split a research plan into a set of coherent, non-overlapping subtasks that can be researched independently by separate agents.

Step 1: Find the research plan

Look for research plan files in the research/ directory (relative to the current working directory) using the pattern research/**/*-research-plan.md.

• If exactly one plan file is found, use it.
• If multiple plan files are found, list them for the user and ask which one to use.
• If no plan files are found, ask the user to provide the path to the research plan file or paste its contents.

Read the selected plan file and extract the slug from the YAML frontmatter (you will need it in Step 3).

Step 2: Split the plan into subtasks

Break the research plan into subtasks following these rules:

• 3 to 8 subtasks is usually the right range — use your judgment based on scope.
• Each subtask must have:
  • id: a short, unique string (e.g., "background", "regional-trends", "key-actors")
  • title: a short descriptive title
  • description: clear, detailed instructions for the sub-agent covering exactly what to research
• Subtasks must collectively cover the full scope of the original plan without unnecessary duplication.
• Prefer grouping by natural dimensions: time periods, regions, actors, themes, causal mechanisms, etc.
• Each description must be self-contained — the sub-agent will not see the other subtasks or the original plan.
• Do not include a final synthesis or summary task; that step happens separately.

Step 3: Save the subtasks to a file

Save the subtasks as a JSON file at:

research/<slug>/<slug>-subtasks.json

where <slug> matches the slug from the research plan's frontmatter.

File format:

{
  "slug": "<slug>",
  "subtasks": [
    {
      "id": "string",
      "title": "string",
      "description": "string"
    }
  ]
}

After writing the file, tell the user where it was saved and give a brief summary of the subtasks created.