# Agent Permission Rule

> A rule that turns one helper agent on or off.

**Field:** Artificial Intelligence  
**Category:** Claude Code  
**Also known as:** Agent(Explore), Agent(name)

## Think of it like

It is like you tell the boss which helpers they may call in.

## Definition

An Agent permission rule matches the subagents Claude may start. It is written as Agent and the name, such as Agent(Explore). Most of the time it is a deny rule, to turn one off. The flag that blocks tools takes the same form.

## Example

Emma wants Claude to search the code itself, not hand it to a helper. She denies Agent(Explore).

## Example code

```json
{
  "permissions": {
    "deny": ["Agent(Explore)"]
  }
}
```

Turn off the Explore subagent.

## Why it matters

Each subagent spends tokens and works out of sight. A rule chooses which ones may run.

## Common mistake

Writing the subagent's label from the screen. Use its real name, as in its file.

## Three principles

- **Agent and a name.** Agent(Explore).
- **Mostly deny.** Turn one off.
- **Same as the flag.** --disallowedTools.

## Related terms

- [Subagent](https://learn.benibix.com/term/ai/subagent)
- [Built-in Subagents](https://learn.benibix.com/term/ai/built-in-subagents)
- [--disallowedTools Flag](https://learn.benibix.com/term/ai/disallowed-tools-flag)

---

Source: https://learn.benibix.com/term/ai/agent-permission-rule
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: 2026-11-17
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
