> ## Documentation Index
> Fetch the complete documentation index at: https://agent-harness101.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Subagent isolation

> Because exploration pollutes the context used to integrate the fix.

## The question

Why not let the main agent investigate everything?

## The answer

Because exploration pollutes the context used to integrate the fix.

### Without subagents

Auth exploration: 20 files

Test logs: 8,000 lines

Dependency dead ends

User conversation

Final reasoning

### With subagents

**Auth → finding**

**Tests → root cause**

**Deps → constraints**

Main context receives concise results.

* **+90.2%** — task success vs. single agent — Anthropic’s own research system

* **\~15×** — tokens a full multi-agent run uses vs. one chat turn

* **80%** — of that performance gap explained by token spend alone

<Warning>
  These numbers are from Anthropic’s own multi-agent *research* system (Opus lead + Sonnet subagents on breadth-first search tasks) — **not Claude Code specifically**. Anthropic explicitly notes the pattern is **worse** for tightly-coupled work like most coding, which is exactly why the next question gates delegation instead of applying it everywhere.
</Warning>

### Source

[anthropic.com/engineering — built-multi-agent-research-system](https://www.anthropic.com/engineering/built-multi-agent-research-system)

<Tip>
  **Design decision:** Delegate isolated exploration, not simply “large tasks.”
</Tip>

## Continue the walkthrough

<Card title="The delegation gate" icon="arrow-right" href="/cost/the-delegation-gate">
  Continue to topic 10 of 14.
</Card>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.