On 27 October 2026, the SpaceXSI docs release notes listed a new API capability.
You can now set content guidelines on text API requests. Add a
guidelinesarray to a Chat Completions or Responses API request to set how Grok handles content in thensfw,violence, andprofanitycategories. Grok follows the guidelines as it writes its response.
What was announced
From the release notes:
- A
guidelinesarray can be added to Chat Completions or Responses API requests. - The array sets how Grok handles content in the
nsfw,violence, andprofanitycategories. - Grok follows the guidelines as it writes its response.
- Documentation is linked at Custom Moderation Controls.
Context
This is an incremental API control for developers using Grok text models. It follows earlier release notes on models such as Grok 4.7 and voice endpoints.
Limits of this report
The release notes do not describe the exact schema of the guidelines array, default behaviour, which models support it, rate limits, or examples. Only the wording in the release notes is reported.