chore: address Copilot review round 1

- skill-manifest: the closing `---` fence must occupy a whole line. `---oops`
  was accepted as a terminator, silently truncating the frontmatter and passing
  validation — the exact failure the guard exists to catch. Regression cases
  added (prefix, trailing whitespace, EOF, CRLF).
- skill-eval: reject a nonzero `claude -p` exit instead of resolving it as an
  empty observation. An auth error or rate limit was scoring as "no skill
  invoked", which passes every negative case and reads as a trigger miss on
  every positive one.
- skill-eval: extract `collectSkillInvocations` / `sampleHit` behind a `main()`
  guard and cover them in `skill-eval.test.mjs` — the stream parsing and the
  process-outcome handling are unreachable from an eval run's happy path.
- docs/test-servers: annotate the era on every catalogue row. Six legacy rows
  carried no annotation under a "unless noted, Modern" default, so the doc sent
  readers to the wrong era for the feature they were trying to see.
- docs/quality-gate: `validate` runs four guards, not three; add the
  `verify:skills` row and the two new `test:scripts` suites; fix the `#publishing`
  fragment left dangling by the split.
- issue-create / pr-flow: delegate the card steps to `/board-ops` rather than
  restating its option IDs, which are explicitly unstable.
- issue-triage: the audit counted presence, not cardinality — an issue carrying
  both `v1` and `v2` passed a check that claims "exactly one". Same for the five
  type labels. Also scope the two milestone checks to board #28.
- AGENTS.md: codify the v1 milestone exception. Every milestone is a v2 release
  bucket, so a v1 issue has none to take and the Incoming-milestone invariant
  cannot apply to board #11.
- clients/launcher/README: its pointer promised an explanation of why the cli/tui
  manifests ship, which the extracted guide does not make.

Signed-off-by: cliffhall <cliff@futurescale.com>

scripts/lib/skill-manifest.mjs

@@ -49,12 +49,15 @@ export function splitFrontmatter(text) {
     };
   }
   const rest = text.slice(FRONTMATTER_FENCE.length + 1);
-  const end = rest.indexOf("\n" + FRONTMATTER_FENCE);
-  if (end === -1)
+  // The closing fence must occupy the WHOLE line. Matching a mere prefix would
+  // accept `---oops` as a terminator, silently truncating the frontmatter to
+  // whatever preceded it — the very failure this function exists to catch.
+  const close = /\n---[ \t]*(\r?\n|$)/.exec(rest);
+  if (close === null)
     return { error: "frontmatter is never closed by a `---` line" };
   return {
-    frontmatter: rest.slice(0, end),
-    body: rest.slice(end + FRONTMATTER_FENCE.length + 2),
+    frontmatter: rest.slice(0, close.index),
+    body: rest.slice(close.index + close[0].length),
   };
 }
 

scripts/lib/skill-manifest.test.mjs

@@ -28,6 +28,25 @@ test("splitFrontmatter requires the fence on the very first line", () => {
   assert.equal(splitFrontmatter(fm("name: x")).error, undefined);
 });
 
+test("splitFrontmatter requires the closing fence to be a whole line", () => {
+  // `---oops` is not a terminator. Accepting it would silently truncate the
+  // frontmatter to whatever preceded it and report the file as valid.
+  assert.match(
+    splitFrontmatter("---\nname: x\n---oops\nbody\n").error ?? "",
+    /never closed/,
+  );
+  // Trailing whitespace on the fence is still a fence, and so is EOF.
+  assert.equal(
+    splitFrontmatter("---\nname: x\n---  \nbody\n").frontmatter,
+    "name: x",
+  );
+  assert.equal(splitFrontmatter("---\nname: x\n---").frontmatter, "name: x");
+  assert.equal(
+    splitFrontmatter("---\nname: x\n---\r\nbody").frontmatter,
+    "name: x",
+  );
+});
+
 test("splitFrontmatter reports an unterminated block", () => {
   assert.match(splitFrontmatter("---\nname: x\n").error ?? "", /never closed/);
 });