Unclear It does not say what makes a Claude Code process a runner child, so it is unclear which setups now lose a mode from settings.
What
The permission mode controls how much Claude Code may do without asking you first. The message that changes it, the set_permission_mode control request, has changed in two ways:
- It can now include
from_settings, marking a mode that came from settings rather than from the user. The code that builds this request now takes a set of options instead of a single positionalultraplanargument. - A runner child, a Claude Code process started and controlled by another one, ignores a
set_permission_moderequest that hasfrom_settingsset. It replies with the mode it already has.
Why
A permission mode supplied by settings can no longer override the mode a runner child is already using. If you rely on settings to set the mode, it may not take effect in these child processes.
The entry above is what we published on the day. These lines were added later, as Anthropic's own pages caught up, and they sit beside the original rather than replacing it.
* **To return to auto mode**: switch to a supported model, then call `set_permission_mode("auto")` or `setPermissionMode("auto")`. If you switch the model back without making that call, the session stays in `default` mode. While the unsupp…agent-sdk/permissions see the edit
It does not say what makes a Claude Code process a runner child, so it is unclear which setups now lose a mode from settings.
Anthropic's documentation has since written up set_permission_mode, on Configure permissions.