[JarryShaw]

• Searched for similar pull requests
• Followed the coding style
• make test passes, and a test case covers the change (narrow selections; see below)
• 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 — workflows or build tooling

---

Description

Closes #2140 (item 2; item 1 was finished by the publish dispatch in #2060).

• Problem. Create Release's "vX is unfinished (fix(ci): Create Release tags main's tip at approval time, not the commit it was created for #1634)" warning, from report_elsewhere() in util/release_status.py, always advised re-running the tagged commit's Create Release run. That re-run repeats the tag's own conda legs, so it cannot upload files those legs could not build. 1.5.0b14's osx-64 legs were broken at its tag.
• Fix. The existing sentence is kept verbatim, because the conda tag and PyPI files still need the re-run. When PCAPKIT_CONDA_COMPLETE != 'true', one sentence is appended. It names Conda Update's publish dispatch on main (ci(conda): add a publish input to Conda Update's manual run (#2060) #2150) for the Anaconda files, as long as main's version is still the tagged one.
• Docs. releasing.rst explains the remedy under the one-commit-per-release rule and in Recovery.
• Tests. test_release_gates.py updates the full-text pin and adds a test for each of the Anaconda-missing and PyPI-only cases. test_release_status_scripts_1967.py strips the new sentence before comparing against the shell copy. The new and updated tests fail on origin/main (2 failed). Selections: 250 + 96 + 9 + 10 passed.

[JarryShaw]

GOOD TO GO at 1dccf8284 (Sonnet review).

• The remedy is true. anaconda_source in util/release_published.py counts any build of the version, so build 1 counts as well as build 0, and a publish dispatch's upload clears completeness. "While main's version is still {version}" is the right condition, since the dispatch uploads main's version.
• The new sentence is gated correctly. Its condition, PCAPKIT_CONDA_COMPLETE != 'true', is the same one that adds Anaconda to missing. The original sentence is byte-identical, and a PyPI-only miss gets no new text.
• The 1967 agreement test still compares the script with the shell copy. Its single-line ANACONDA_REMEDY strip hides nothing else.
• Tests: 10 + 9 + 133 passed. The new tests fail 2 on origin/main.
• Doc nit, non-blocking: "names that remedy wherever Anaconda files are missing" is slightly broader than the warning's real trigger (a version tagged at another commit).

[github-actions]

Coverage: 91.55% (unit tier, Python 3.15, 1dccf8284, 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	2589	75	816	37	96.48%
pcapkit/dumpkit	282	3	102	3	98.44%
pcapkit/foundation	2971	23	1036	25	98.80%
pcapkit/interface	138	4	52	3	96.32%
pcapkit/protocols	17418	219	4574	185	98.14%
pcapkit/toolkit	869	36	298	10	95.20%
pcapkit/utilities	504	7	146	5	98.15%
pcapkit/vendor	4432	1988	1018	167	51.87%

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

[JarryShaw]

Ready to merge at 1dccf8284.

• Review: Sonnet GOOD TO GO.
• CI: 61 CheckRuns passed, 0 failed, 0 running, and Required checks passed is SUCCESS. The KTLO legs show no warnings.
• Ordering: tests/project on this PR merged into main (293d3baa6) passes the plain unittest ordering leg, with 2288 tests, 0 failures and 0 errors.
• Merging closes release: 1.5.0b14 has no osx-64 conda files, so Create Release warns it is unfinished on every released-elsewhere run #2140. Item 1 is already confirmed on main by Create Release 40152325584.