A new SDK control-protocol handler, update_settings, lets clients update the outputStyle setting only, requiring string values and disallowing deletion. It only works when the localSettings source is enabled for the session (see --setting-sources) and not over a remote transport.
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.
Added since
A small documentation edit on Orchestrate teams of Claude Code sessions touched a line naming --setting-sources after this was published.
Each teammate has its own context window. When spawned, a teammate loads the same project context as a regular session: CLAUDE.md, MCP servers, and skills. If you start the lead with [`--setting-sources`](/docs/en/cli-reference#cli-flags),…agent-teams see the edit
Confirmed since
Anthropic's documentation has since written up --setting-sources, on Extend Claude with skills.
* A session where your organization's managed settings [lock skills to plugin sources](/docs/en/settings-reference#strictpluginonlycustomization-skills), or one you start with a [`--setting-sources`](/docs/en/cli-reference#cli-flags) list …skills see the edit
Added since
A small documentation edit on Modifying system prompts touched a line naming outputStyle after this was published.
In the Python SDK, set `outputStyle` through the `settings` option, which takes a JSON string such as `'{"outputStyle": "Explanatory"}'` or a path to a settings file that sets it.agent-sdk/modifying-system-prompts see the edit
Confirmed since
Anthropic's documentation has since written up outputStyle, on How Claude Code uses prompt caching.
When you switch [output styles](/docs/en/output-styles) mid-session with `/config` or the `outputStyle` setting, Claude uses the new style starting with your next message. Claude Code delivers the new style's instructions as a message in t…prompt-caching see the edit
One source agreesOne thing we can check says the same as this entry.
Anthropic's documentation agrees
Anthropic's documentation has since written up outputStyle, on How Claude Code uses prompt caching.