Split a big task across custom subagents

Split a big task across custom subagents

AI & Agents
Anthropic
A researcher, a writer, and a fact-checker each with its own prompt and permissions, instead of one prompt trying to be all three.
Source
Anthropic
Tools
Claude Code
Runs
On demand
Cost
Free, model usage applies
Last updated
August 7, 2026

What it is.

A subagent is a file describing one worker: what it does, which tools it may use, and which model it runs on. You choose the scope, so a subagent can be personal to you or shared with a project.

The value for marketing work is separation. A research subagent that only reads, a drafting subagent that only writes, and a checking subagent that verifies claims each stay small enough to behave predictably.

You also control capabilities per subagent, which is how you stop a drafting worker from touching anything it should not.

What you get.

  • Named subagents, each with its own instructions and tool permissions.
  • A model choice per subagent, so cheap work runs on a cheap model.
  • Project or personal scope, so a team can share the same set.
  • Capability limits per subagent, which keeps a writer from editing files.
  • Parallel work on one task, split across focused workers.
HOW TO USE IT

How to set it up.

1

Write down the task as separate jobs before creating anything, since the split is the whole design.

2

Create one subagent per job and describe what it does and when to use it.

3

Choose the scope, with project scope when the team should share it.

4

Limit each subagent's tools to what its job needs.

5

Pick the model per subagent, keeping the expensive one for the hardest step.

6

Test each subagent alone before chaining them, because a bad handoff is hard to debug in a chain.

Use cases

Split research from writing

Give each subagent its own instructions and let them work in sequence.

Add a checker

Point one subagent at verification so the writer is not reviewing itself.

Spend the model budget wisely

Use a cheaper model on the mechanical steps and keep the expensive one for judgment.

Best for

Research then write then check

Three jobs with different tools and standards, which is exactly what one prompt handles badly.

Cost control

A cheap model can do extraction while the expensive one only handles judgment.

Shared team setups

Project scope means everyone delegates to the same workers.

Open the original.

Hosted on Anthropic, free to open.
Open the Tool

Questions about Split a big task across custom subagents

When is fanning out worth it?
Where does quality get lost?
Why is delegation unpredictable?

Questions about AI & Agents

How many agents should a task be split across?
What permissions should a subagent get?
Strengths
  • Tools and model are set per subagent, so cost and risk are controlled at the right level.
  • Scope is explicit, so a personal experiment does not leak into the team's setup.
  • The docs cover capability limits, which is what makes delegating to a writer safe.
Limitations
  • Too many subagents with overlapping descriptions makes delegation unpredictable.
  • Each subagent adds its own model usage, so fanning out costs more than one pass.
  • Handoffs between subagents are where quality is lost, so check the intermediate output.
  • A subagent inherits whatever its tools can reach, so review permissions before sharing it.
Skip this if
  • Skip it if your work fits in one pass and needs no separate reviewer.
Ideal for
  • Marketing ops leads
  • Agency operators
  • Stage: Any stage

Want this running without building it yourself?

TripleDart has scaled 300+ tech companies with expert operators and AI workflows behind every play.