chore(devx): migrate the 7 freed scripts/** line citations to file-level anchors (#15809, residual 2 of 3)

scripts/check-docs-section-name.mjs

@@ -1470,9 +1470,10 @@ export function selfTest() {
     t('a NAMED YAML mapping satisfies the rule', run(yamlNamed, quiet), EXIT_CLEAN);
 
     // ── ⭐ #13880 selector ③: a singular `section:` mapping, nameless ───────
-    // The exact shape `concept.mdx:426` carried live: `section:` (not
-    // `sections:`) whose value is a mapping, nested under an unrelated outer
-    // key -- the `- section:` sequence-item cardinality.
+    // The exact shape `content/docs/protocol/objectui/concept.mdx` carried
+    // live (line 426 as measured): `section:` (not `sections:`) whose value
+    // is a mapping, nested under an unrelated outer key -- the `- section:`
+    // sequence-item cardinality.
     battery('⭐ #13880 selector ③: a singular `section:` mapping, nameless');
     const yamlSingularNameless = tree(
       baseFixtureFiles({
@@ -1508,8 +1509,9 @@ export function selfTest() {
 
     // ── negative controls: the widened `sections?:` prefilter must not ──────
     // fabricate a population out of `_section:`, `.section` member access, or
-    // a scalar `section:` value (no mapping to judge -- the doc-pages.mdx:257
-    // control this card's own triage named: "...from the previous section:"
+    // a scalar `section:` value (no mapping to judge -- the control at
+    // `content/docs/ui/doc-pages.mdx` (line 257 as measured) this card's own
+    // triage named: "...from the previous section:"
     // is prose outside any fence and never reaches this regex at all).
     battery('negative controls: the widened `sections?:` prefilter must not');
     const yamlSingularNegatives = tree(
@@ -1564,7 +1566,8 @@ export function selfTest() {
     t('a `jsonc` fence with comments and a trailing comma is judged, not refused', run(jsonWithComments, quiet), EXIT_CLEAN);
 
     // ── an ELIDED placeholder array: judged as ZERO entries, not skipped ────
-    // (`forms.mdx:183`'s real shape: `"sections": [/* … */]`)
+    // (`content/docs/ui/forms.mdx`'s real shape, line 183 as measured:
+    // `"sections": [/* … */]`)
     battery('an ELIDED placeholder array: judged as ZERO entries, not skipped');
     const jsonElided = tree(
       baseFixtureFiles({

scripts/check-scripts-symbol-anchors.mjs

@@ -72,11 +72,17 @@
  * repointed; every number that was in an anchor is still on its page, as data.
  *
  * The residual when that landed was 15, and every one of them was a file
- * another lane held OPEN at that moment -- ⛔ not one was an ambiguity. That
- * count is a DATED reading; `--list-unresolvable` is the live one. ⚠️ It is why the
+ * another lane held OPEN at that moment -- ⛔ not one was an ambiguity. A
+ * second round (#15809, still) migrated the 7 of those 15 whose holders had
+ * by then merged -- `check-docs-section-name.mjs` and
+ * `docs-audit/affected-docs.mjs` (3 each) and `check-type-check-coverage.mjs`
+ * (1) -- leaving 8, again every one held OPEN and none an ambiguity: 4 in
+ * `check-system-context-census.mjs` and 3 in `cross-package-test-inputs.mjs`
+ * (PR #16215), 1 in `pm/check-half-states.mjs` (PR #16202). Both counts are
+ * DATED readings; `--list-unresolvable` is the live one. ⚠️ It is why the
  * flag is still `false`: the fence #15809 was dispatched under is that it flips
  * only when the residual is ZERO and a self-test pins the flip, and a residual
- * of 15 would make this gate permanently red for the length of somebody else's
+ * of 8 would make this gate permanently red for the length of somebody else's
  * pull request. `--list-unresolvable` prints the residual so the next author
  * inherits a worklist rather than a count; the day it prints nothing, the flag
  * is a one-line change with a case to pin it.

scripts/check-type-check-coverage.mjs

@@ -350,8 +350,10 @@
 //               about what that costs in practice.
 //
 //               What drifts is not only the NUMBER but the note's COMPOSITION:
-//               service-automation's note named `engine.test.ts:2547/2577` as
-//               the whole debt while three TS2341 in a different file, from an
+//               service-automation's note named its
+//               `packages/services/service-automation/src/engine.test.ts`
+//               (lines 2547/2577 as measured) as the whole debt while three
+//               TS2341 in a different file, from an
 //               unrelated PR, had joined it. So when this invariant makes you
 //               raise a count, rewrite the note to match what the pile is now
 //               made of -- and when the delta cannot be attributed, say so in

scripts/docs-audit/affected-docs.mjs

@@ -4696,10 +4696,11 @@ function selfTest() {
   // an English comment inside a handler window bridged just the same.
   //
   // MEASURED FAILURE (40d5b2d4c, #9405 — a `metadata-protocol` batch-publish change): BOTH
-  // route anchors that run produced were prose and nothing else.
+  // route anchors that run produced were prose and nothing else, both in
+  // `packages/rest/src/rest-server.ts` (lines as measured then, kept as data):
   //
-  //   promoteDraftForPublish → /:type/:name/publish   rest-server.ts:5324,5376  (comments)
-  //   publishPackageDrafts   → /:name/state/:field    rest-server.ts:5694,5722  (comments)
+  //   promoteDraftForPublish → /:type/:name/publish   lines 5324,5376  (comments)
+  //   publishPackageDrafts   → /:name/state/:field    lines 5694,5722  (comments)
   //
   // The second one carried `content/docs/protocol/objectql/state-machine.mdx` onto the
   // advisory (and `meta.getLegalNextStates` through the ledger) for a diff that went
@@ -4773,7 +4774,8 @@ function selfTest() {
   // The third layer of the same family, and the only one with no measured wrong row on the
   // tree that filed it — read that as the point of the block, not as a reason to skip it.
   //
-  // MECHANISM (verified on e7daea169). `rest-server.ts:5661` registers
+  // MECHANISM (verified on e7daea169). `packages/rest/src/rest-server.ts`
+  // (line 5661 as measured then) registers
   // `/:name/state/:field`; the next LITERAL `path:` is 255 lines later at 5916, so the
   // window runs its full 150 lines to 5810 — straight over `path: publishedPath` at 5747,
   // which registers a different route the scan cannot see. 64 lines of the `published`