Release Notes

Release Notes

Every Core dna product, release by release, newest first. Each product ships on its own schedule.

Platform 2026.5.2

What’s new in Core dna Platform 2026.5.2

Custom Entities and Hooks APIs

Custom entities and jobs can be managed end to end through the API: definitions, records, triggers, actions and run history. Record and run listings are now paged.

Key highlights

  • Custom entities through the API: create and change definitions, activate revisions, duplicate records, and delete or publish records in bulk.
  • Jobs through the API: build, run and inspect jobs, with their triggers, actions and run history.

Changes in behaviour

Some existing API responses changed with this work, and existing integrations should be reviewed against them. Record and batch listings are now paged, returning 100 items by default up to a maximum of 500. A site sent in a record body is ignored, and the record belongs to the site the request addresses. Reading a single record by id now returns unpublished records to administration and integration keys. Batch responses return the hook, site and batch event as objects rather than as bare ids. Payload keys are now matched without regard to case, so a capitalised key for a protected field such as a status is dropped rather than applied. The expiry flag on file payloads was inverted, and now reports whether a file has already expired rather than whether it is set to expire.

Custom entities through the API

Custom entities can now be managed end to end through the API. Alongside the record endpoints, a definition itself can be created, updated and deleted, a record revision activated or removed, a record duplicated, and records deleted or published in bulk, up to 50 at a time.

Deleting a definition that has child definitions or records reports how many would go with it, and goes ahead only when the request confirms the cascade.

Jobs through the API

Jobs, still under /cdna-api/hooks, are now available through the API too. Jobs, categories, record and web event triggers, actions and conditions can each be created, read, updated, deleted and listed; a job can be run on demand; a run can be queued again; and the history of runs and their actions can be read. A job created this way stays unpublished, and does nothing, until it is published, as in the admin.

Run job 42 now 
POST /cdna-api/hooks/hook/42/run HTTP/1.1
Host: hub.coredna.com
Authorization: Bearer YOUR_ACCESS_TOKEN
Accept: application/json

Other improvements and bug fixes

  • Fixed an issue where an API request failed with a server error when a record referenced an unpublished item, such as a product whose primary category is unpublished. Hidden references now come back empty, and saving a reference to a record you cannot see returns a validation message against the field instead of a server error.
  • Fixed an issue where a site search that could not be completed returned a server error even though the page rendered normally. A failed search now shows the search page with no results, and a request for a page beyond the last page of results returns a not-found page.
  • Fixed an issue where adding a sort parameter to a public listing page, such as a product or blog listing, returned a server error instead of the sorted results. An unrecognised sort now falls back to the listing's configured default.
  • Corrected an issue where month labels on the billable requests screen could be shifted back by one month, with the current month missing from the chart, for sites or viewers in a timezone behind UTC.
Go deeper
GuideCustom endpoints with synchronous jobs ReferenceCustom entities API ReferenceJobs API How-toDebug a job run that failed halfway

Was this page helpful?

Your answer goes to the docs team.