# allowDangerouslySkipPermissions Option

> The option that must be true before bypass mode can run.

**Field:** Artificial Intelligence  
**Category:** Claude Agent SDK  
**Also known as:** skip permissions safety flag, Claude Agent SDK allowDangerouslySkipPermissions

## Think of it like

It is like a lock that needs two keys to open.

## Definition

allowDangerouslySkipPermissions is an option in the TypeScript Agent SDK. It must be true before a session can use bypass mode. You need it at the start, and also before you switch to that mode later. The default is false. The long name is there so no one turns on the riskiest mode by mistake.

## Example

Leo sets bypass mode in his CI agent and gets an error. He sets the long safety option to true, and it starts.

## Why it matters

Bypass mode runs tool calls with no checks, so it should take two clear steps to turn on.

## Common mistake

Turning it on outside a sandbox. Keep bypass mode for CI and throwaway containers.

## Three principles

- **Needed for bypass.** In TypeScript.
- **Off by default.** false.
- **Two keys on purpose.** No accidents.

## Related terms

- [permissionMode Option](https://learn.benibix.com/term/ai/permission-mode-option)
- [Bypass Permissions Mode](https://learn.benibix.com/term/ai/bypass-permissions-mode)
- [--allow-dangerously-skip-permissions Flag](https://learn.benibix.com/term/ai/allow-dangerously-skip-permissions-flag)

---

Source: https://learn.benibix.com/term/ai/allow-dangerously-skip-permissions-option
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: 2026-11-20
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
