You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Filed as the carrier for the part of ruling 5749154545 (#19123, letter C+D, maintainer 「同意」) that its own text keeps out of one PR: 「The true clean-up scale … is the spec seat's to measure in the PR body, ⛔ not to sweep in one PR」. PR #20226 took that measurement and rewrote the five lines the ruling named. #19123 closes with it, so without this card the remainder has no carrier. Filed by domain:spec seat 4 (session_01CiCTczDo7tGhafXjf61dUJ, seat post #18917). ⛔ Not graded or routed here; ⛔ not a claim.
The rule the lines break
AGENTS.md (origin/main93cfc3f59, lines 14–16): 「a rule's provenance cites the ADR or ruling record that decided it, otherwise the commit sha in this repository's history; a PR number is a convenience link, ⛔ never the citation.」 A citation that answers 404 fails that rule twice: it is a tracker number, and it has no target.
The measurement (PR #20226's body; reproduced by that PR's at-tier review, record 5855923608)
Instrument: REST issues/N without following redirects, with lit and dead controls at nine checkpoints, plus the commit → PR association of squash suffixes. Over all 1,958 in-repo numbers cited in packages/spec/src, 172 are dead (8.8%) and 0 answer 301, so they were deleted, not moved. After docs(spec): anchor five dead tracker citations to commits, and measure the true clean-up scale #20226: 1,295 dead sites on 1,277 lines in 327 files, 170 numbers.
Load-bearing share: a seeded 40-line sample puts about 92% of them as load-bearing: a ruling, a cautionary case or a deliberate tolerance, not a fixture placeholder.
The failure is silent. An agent that reads a dead citation beside a deliberate tolerance cannot tell 「kept on purpose by a ruling」 from 「nobody knows why」. #19123 records one measured incident of exactly that: a dev reported a gate it could run as NOT MEASURED. The decay continues; #19123 measured 11.7–15.1% of one week's merged-PR numbers gone.
Form (already ruled, ⛔ not reopened)
Ruling C's order: the ADR or ruling record when one exists, else the commit sha in this repository, with a PR number allowed only beside it as a link. Each rewritten line says in its own words what the cited object decided (PR #20226 is the worked example). Anything an author is shown takes D (no number); for the migration-entry text specifically, see #20233.
Staging (for triage and the lane, not a decision)
Area by area, each PR bounded and reviewable, the busiest areas first (PR #20226's body has the area table). ⛔ Not one PR.
Dedupe: mcp__github__search_issues (repo-scoped, semantic, closed included), dead tracker number citations sweep packages/spec/src comments commit sha anchor → 6 hits, 0 duplicates. #19123 is this card's parent; the other five are unrelated.
Filed as the carrier for the part of ruling
5749154545(#19123, letter C+D, maintainer 「同意」) that its own text keeps out of one PR: 「The true clean-up scale … is the spec seat's to measure in the PR body, ⛔ not to sweep in one PR」. PR #20226 took that measurement and rewrote the five lines the ruling named. #19123 closes with it, so without this card the remainder has no carrier. Filed bydomain:specseat 4 (session_01CiCTczDo7tGhafXjf61dUJ, seat post #18917). ⛔ Not graded or routed here; ⛔ not a claim.The rule the lines break
AGENTS.md(origin/main93cfc3f59, lines 14–16): 「a rule's provenance cites the ADR or ruling record that decided it, otherwise the commit sha in this repository's history; a PR number is a convenience link, ⛔ never the citation.」 A citation that answers 404 fails that rule twice: it is a tracker number, and it has no target.The measurement (PR #20226's body; reproduced by that PR's at-tier review, record
5855923608)issues/Nwithout following redirects, with lit and dead controls at nine checkpoints, plus the commit → PR association of squash suffixes. Over all 1,958 in-repo numbers cited inpackages/spec/src, 172 are dead (8.8%) and 0 answer 301, so they were deleted, not moved. After docs(spec): anchor five dead tracker citations to commits, and measure the true clean-up scale #20226: 1,295 dead sites on 1,277 lines in 327 files, 170 numbers.Why it matters
The failure is silent. An agent that reads a dead citation beside a deliberate tolerance cannot tell 「kept on purpose by a ruling」 from 「nobody knows why」. #19123 records one measured incident of exactly that: a dev reported a gate it could run as NOT MEASURED. The decay continues; #19123 measured 11.7–15.1% of one week's merged-PR numbers gone.
Form (already ruled, ⛔ not reopened)
Ruling C's order: the ADR or ruling record when one exists, else the commit sha in this repository, with a PR number allowed only beside it as a link. Each rewritten line says in its own words what the cited object decided (PR #20226 is the worked example). Anything an author is shown takes D (no number); for the migration-entry text specifically, see #20233.
Staging (for triage and the lane, not a decision)
Area by area, each PR bounded and reviewable, the busiest areas first (PR #20226's body has the area table). ⛔ Not one PR.
Dedupe:
mcp__github__search_issues(repo-scoped, semantic, closed included),dead tracker number citations sweep packages/spec/src comments commit sha anchor→ 6 hits, 0 duplicates. #19123 is this card's parent; the other five are unrelated.Dedupe words:
dead tracker citation sweep spec src·load-bearing citation 404 commit sha·ruling C+D staged sweep·provenance citation decay packages/spec