# SubagentStop Hook

> The event when a helper agent ends. A hook can make it go on.

**Field:** Artificial Intelligence  
**Category:** Claude Code  
**Also known as:** SubagentStop

## Think of it like

It is like you check a helper's work. Then they can go home.

## Definition

SubagentStop is the hook event that fires when a subagent finishes its answer. It works like Stop, but for the helper. A hook can block the stop with a reason, and the subagent keeps working. It also fires for small agents Claude Code runs for itself, which have an empty agent name.

## Example

Jack's review helper kept ending with no summary. His hook blocks it until a summary is there.

## Example code

```json
{
  "decision": "block",
  "reason": "List the files you changed before you finish."
}
```

Send the subagent back with a task.

## Why it matters

A subagent's answer is all the main chat gets. This checks it is complete.

## Common mistake

Using an empty matcher and being surprised by extra runs. Internal agents fire it too.

## Three principles

- **When a helper finishes.** Like Stop.
- **Block with a reason.** It goes on.
- **Internal agents too.** Empty name.

## Related terms

- [SubagentStart Hook](https://learn.benibix.com/term/ai/subagentstart-hook)
- [Stop Hook](https://learn.benibix.com/term/ai/stop-hook)
- [Subagent](https://learn.benibix.com/term/ai/subagent)

---

Source: https://learn.benibix.com/term/ai/subagentstop-hook
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-16
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
