Aggregations section in GraphQL API documentation

[meganelacheny]

• Addition of an "Aggregation" section in main GraphQL API page
• Addition on an aggregation example in GraphQL - Advanced Queries page

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Preview	Comments	Updated (UTC)
documentation	Ready	Preview	Comment	Nov 18, 2025 10:55am

[pwizla]

This is a beautiful PR, Mégane, thank you! #chefkiss

The only minor comment I have is that I think the {#aggregations} part that follows the header on line 428 of graphql.md is unnecessary 😊

## My title {#my-custom-heading-title} is the format for custom heading ids that we sometime use when we want to manually define a precise heading link. This is useful for instance when we add badges to titles, to ensure the badge text is not part of the anchor link.

Here, since the ## Aggregations h2 and its custom heading {#aggregations} are strictly identical, I suggest we remove it; Docusaurus will properly generate a /cms/api/graphql#aggregations anchor automatically 🤓

[pwizla]

Oh, and I've just noticed this 👀

Suggestion to replace the link by a sentence about depthLimit and amountLimit.

The rate-limiting-and-query-depth section does not exist in our docs 😅 causing the build to fail (broken anchor link)

Suggested change

Aggregations operate server-side, so they are generally faster than downloading and processing large result sets on the client. However, complex `groupBy` trees and wide projections can still be expensive. Use filters to restrict the data set and consider enabling [query depth or amount limits](/cms/plugins/graphql#rate-limiting-and-query-depth) to protect your API. Errors such as `You are not allowed to perform this action` usually mean the requester lacks the `Read` permission on the target collection.

Aggregations operate server-side, so they are generally faster than downloading and processing large result sets on the client. However, complex `groupBy` trees and wide projections can still be expensive. Use filters to restrict the data set and consider setting up `depthLimit` and `amountLimit` values accordingly (see [#available-options](#available-options)) to protect your API. Errors such as `You are not allowed to perform this action` usually mean the requester lacks the `Read` permission on the target collection.

Information from Kapa since I didn't know whether we had query limiting options for GQL in Strapi 😅

[meganelacheny]

Ah oops, well spotted 🙈 Thank you for the proposition to fix that mistake!