What happened? What is the problem?
Using /backport skill and if verified ways of bumping spec fail (f.e. rpmdev-bumpspec fails because the spec is old - f.e. has old syntax for applying patches, or date was not called), it creates the date in spec changelog manually and creates bogus date.
Note: if rpmdev-bumpspec -c 'mesage' *.spec (or its rightmost bump version) fails, the release number is bumped correctly and changelog message is almost correct - the only thing missing in spec changelog is NVR.
log3.txt
What did you expect to happen?
Generate correct date at first try.
Example URL(s)
No response
Steps to reproduce
1. start Claude Code v2.1.145, opus 4.6
2. /backport CVE-2026-35177 fix to rhel-9.8.0 and c8s
3. get into backport to rhel-9.8.0, where rpmdev-bumpspec fails - claude will try to guess the date by itself, but incorrectly
Workaround
Participation
What happened? What is the problem?
Using /backport skill and if verified ways of bumping spec fail (f.e. rpmdev-bumpspec fails because the spec is old - f.e. has old syntax for applying patches, or
datewas not called), it creates the date in spec changelog manually and creates bogus date.Note: if
rpmdev-bumpspec -c 'mesage' *.spec(or its rightmost bump version) fails, the release number is bumped correctly and changelog message is almost correct - the only thing missing in spec changelog is NVR.log3.txt
What did you expect to happen?
Generate correct date at first try.
Example URL(s)
No response
Steps to reproduce
Workaround
Participation