This release updates our documentation to explain a rare release hook timeout scenario and what users should expect if their client retries a request.
Clarified behavior for release hook timeouts
We documented an edge case where a release hook request can appear to time out from the caller’s perspective, even though the original request may still complete successfully. In this situation, a retry can overlap with the in-flight request and previously could result in duplicate published entries for the same version.
The notes now describe the expected retry outcome, including that overlapping retries should wait and then receive a conflict response rather than creating a second entry. They also clarify how “skip changelog” decisions are handled when the initial check takes too long, and how usage is handled if an entry is later removed because everything was skipped.