16i-path-citations: point the clone URL at the new org, and gate dead links - #12
Merged
Conversation
… links
The repository moves to the ScriptKittyOS organisation. GitHub redirects the
old URL, so a stale citation keeps working while hiding that it is stale --
which is why the link check lands with the citation and before the transfer,
not after. Run after, it would pass on a tree still full of old paths.
The population was derived by grep at the SHA it runs against, not taken from
the list in the transfer issue. That issue names six files; the derivation
returns one:
git grep -in 'HackTuah/Ultraviolet' 778accb --
README.md:209 files: 1 occurrences: 1
NOTICE carries no URL, CITATION.cff does not exist yet, schemas/README.md and
CHANGELOG.md have never been on main, and nothing under docs/ cites the path.
The link check fails on any dead relative link in README.md or docs/, guarding
27 links across 14 markdown files. It carries four probes, and each is
demonstrated able to fail rather than merely observed passing:
neuter dead_links/2 to [] probes 1,2,3 RED, probe 4 GREEN
append a real dead link to README probe 4 RED
Probe 4 staying green under a blind checker is the point, not an oversight: a
checker that enumerates nothing reports zero dead links and passes. Probes 1-3
are what catch it. dead_links/2 takes its root and source list as arguments so
the probes drive the same function the real test calls; a control that computes
its expected value with the function under test proves nothing.
Both reviewers reproduced the tree independently and confirmed the mutations in
their own copies. Suite 260 tests, 0 failures -- 254 at 268a83e, +2 from 16g,
+4 probes, each leg re-derived by review.
Acceptance criteria 1-6 met; 7 is the PR's nine contexts. Evidence in
internal/slices/16i-path-citations/FINDINGS.md, which carries an appended
correction of two numbers that did not reproduce.
The GitHub transfer itself is owner work, after this merges. Between merge and
transfer, main ships a clone URL naming an org that does not yet own the repo:
an accepted consequence of the ordering above.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VNdY6wVS773UJgrHUeBN3Y
Reviewed-diff: sha256:93670cbcc3e772e32419e4e2839a6942fd235fef64ac85e12a0632bcd8fcc286
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Splits the path-citation half out of the transfer bundle so it can land before the org move, and brings the link-check with it.
Why before, not after
GitHub redirects the old URL. A stale citation keeps working while hiding that it is stale, so a link check run after the transfer passes on a tree still full of old paths and proves nothing. Run before, it fails on exactly what the transfer invalidates.
The population was derived, not listed
The transfer issue names six files. The derivation returns one —
NOTICEhas no URL,CITATION.cffdoes not exist yet,schemas/README.mdandCHANGELOG.mdhave never been onmain, nothing underdocs/cites the path. That issue's own instruction was to derive rather than work from its list; the instruction was right and the list was wrong.The link check, and why its probes are evidence
Guards 27 relative links across 14 markdown files. Four probes, each demonstrated able to fail:
Probe 4 staying green under a blind checker is the finding, not an oversight: a checker that enumerates nothing reports zero dead links and passes. Probes 1–3 are what catch it.
Review
Two lanes, both PASS, both reproducing the tree independently and re-running the mutations in their own copies. Suite 260 tests, 0 failures — 254 at
268a83e, +2 from 16g, +4 probes, each leg re-derived by review. Every hook gate green including the signoff and reviewed-tree binding.Residuals ruled residual by the owner and filed as two derivation-only issues: the enumeration-side evidence gap, and three checker limits with zero live instances.
Owner work after merge
The GitHub transfer itself. Between this merging and the transfer completing,
mainships a clone URL naming an org that does not yet own the repository — an accepted consequence of the ordering above.🤖 Generated with Claude Code
https://claude.ai/code/session_01VNdY6wVS773UJgrHUeBN3Y