[JarryShaw]

• Searched for similar pull requests
• Followed the coding style (make pylint, make mypy, make isort)
• make test passes, and a test case covers the change
• Added a changelog entry — N/A — centralised in docs(changelog): shared 1.5.0 changelog — long-lived, merges last (#610, #616, #617, #618, #620) #657

What is the purpose of your pull request?

• ci — changes to CI configuration files and scripts

---

Description of your pull request and other information

This is the last create-release batch for #1967. Each step of the github and tag jobs, and cron-conda's Check the GitHub release exists, was judged against the inline rule.

Step	Lines	Verdict
github / Select release body	15	keep inline: one place, never fixed
github / Explain a refused release tag	6	keep inline: trivial
tag / Create Tag	20	keep inline: one place. Its overlap with release_tags.py reads the tag differently.
tag / Reset Build Number	44	move: a 3-attempt push loop with 5 exits; it is the #1634 fix
cron-conda / Check the GitHub release exists	18	keep inline: shares no logic with create-release

• util/release_reset_build.py: stdlib only and 3.7-parseable, since the tag job has no setup-python. The step keeps its name, env and position.
• Equivalence: test_release_reset_build_1967 runs the old shell verbatim beside the script, on scratch repos. It covers 24 cases across every exit, comparing status, output, origin refs, HEAD and conda/build. Five planted divergences were caught.
• Results: the release, tag, helper and actionlint tests pass.

[JarryShaw]

GOOD TO GO at eabadbdab (Sonnet review of 3840ba814, plus a test-only amend).

• Behaviour: the review found no divergence from the old step. Retries, tag parse, conda/build strip, commit/switch/push arguments, exit codes and messages all match. There is no pull, merge, reset or force push. The PyYAML diff touches only the step's run:.
• Equivalence: FORMER_RESET equals the old body exactly. Every case reaches its expected exit, and four of the reviewer's own planted mutations failed.
• Amend: the review found that dropping --force from git switch passed every test. eabadbdab now pins it in the git-call test and adds a dirty-checkout case. Without --force, those 2 tests fail; with it, 7 pass with 97 subtests.
• Create Tag: the reviewer argued it could move (20 lines, three-way branch, fix(ci): Create Release tags main's tip at approval time, not the commit it was created for #1634). It stays inline, like vendor_update's 19-line single-site step under the same rule. That call is open to the owner.

[github-actions]

Coverage: 91.54% (unit tier, Python 3.15, eabadbdab, Unit Tests run success)

Package	Statements	Missed	Branches	Partial	Cover
pcapkit (top level)	104	4	20	2	95.16%
pcapkit/const	19599	880	2618	866	92.12%
pcapkit/corekit	2565	71	812	37	96.56%
pcapkit/dumpkit	282	3	102	3	98.44%
pcapkit/foundation	2930	23	1022	25	98.79%
pcapkit/interface	138	4	52	3	96.32%
pcapkit/protocols	17414	219	4574	185	98.14%
pcapkit/toolkit	869	36	298	10	95.20%
pcapkit/utilities	473	7	136	5	98.03%
pcapkit/vendor	4432	1988	1018	167	51.87%

Per-file detail: the coverage-html artifact of this run.

[JarryShaw]

Ready to merge at eabadbdab.

• Review: the Sonnet review is GOOD TO GO, with the --force test gap closed in the amend.
• CI: every blocking check passed (Required checks passed SUCCESS). The 3.8–3.10 KTLO legs carry no warnings.
• Ordering: the tests/project plain-unittest ordering leg passed on a scratch merge into main, 2121 tests with 0 failures.
• Base: the head is on current main (cfabfb2d3).

With this merged, every create-release step has been judged under the #1967 inline rule.