fix(#4636,#4653): close the symlink hole, revert a wrong collapse, fix six review findings

The RED checkpoint and two orthogonal reviews found eight defects. All fixed here.

THE COLLAPSE THAT WAS WRONG — installer-migrations. Routing ensureInsideConfig's
containment decision through the realpath-based canonical predicate broke four
tests, and the failure message says it plainly: "migration path escapes
configDir: extensions/gsd.cjs". That module's entire contract is that a
symlinked managed path is snapshotted, restored and backed up AS A LINK and
never dereferenced. The canonical predicate dereferences, then rejects the
result for escaping configDir — so it destroys exactly the thing the module
exists to preserve. Reverted to lexical, with the ruling recorded above the
function so it is not collapsed a third time. normalizeRelPath is the real
pre-gate there; it throws on absolute paths and '..' before this check runs.

That makes THREE deliberately-retained implementations, not two, and they share
one shape worth naming: a realpath-based predicate is the wrong tool wherever a
symlink must be PRESERVED rather than resolved. CONTEXT.md and
docs/explanation/security-model.md are corrected — both previously described
ensureInsideConfig as collapsed.

THE MISSED CONSUMER. tests/security-prompt-injection.security.test.cjs
destructures validatePath from the compiled lib; un-exporting it turned five
tests into TypeError. It appeared in my own earlier search output and I did not
follow it up. Translated under the same rule as the rest: assertions on the
rejection REASON go through assertWithinRoot, boolean-only through
tryWithinRoot.

VALIDATE-ONE-PATH-USE-ANOTHER, FOUND TWICE MORE. This is the fourth and fifth
occurrence in this epic of the exact defect it exists to prevent.
  - scripts/check-glossary-refs.cjs decided containment on `token` and then
    stat'd a separately re-joined path.join(ROOT, token). The ContainedPath is
    now carried through to the probe, so the validated value is the probed one.
  - src/init.cts computed skillPathContained and DISCARDED it, re-joining from
    the raw input for the existsSync and read. The branded type exists to make
    that a type error and here it was inert.

AND THE OVER-CORRECTION OF THAT FIX, caught before it shipped. The first attempt
also substituted the validated value into the EMITTED `ref` for a global skill.
That value is a display token, not a path anything reads through — the only fs
access in that branch runs on the lexical path beforehand — so substituting it
changed emitted output two ways: it is realpath-resolved, so a symlinked global
skills directory would have emitted its resolved target instead of the user's
own path, and it came from path.join, so Windows would have emitted a backslash
where the template has a literal '/'. Restored, with the distinction recorded:
the containment check there is a GATE, not a path producer.

A TEST THAT COULD NOT FAIL. The first symlink regression planted its symlink
from inside a hooked fs.readdirSync and never asserted the planting happened —
if the hook did not fire, the "nothing was written outside" assertion passed
trivially, green against vulnerable code. It now asserts the plant, matching its
sibling. The other two were re-checked: one already asserted its equivalent, the
other plants synchronously and cannot silently no-op.

THE SYMLINK FIX ITSELF, now that the tests are proven red on the matrix.
isPathConfined is lexical by design and structurally cannot see a symlink; three
callers relied on it with no defense of their own. install-engine.cts:1608 and
install-profiles.cts:880 refuse to mkdir/write through a link — mkdirSync with
recursive:true does NOT throw on an existing symlink-to-directory, so a planted
link redirected the SKILL.md write outside the install root.
install-profiles.cts:755 refuses to read through one — statSync FOLLOWS links,
so an outside file's contents were returned and installed as a skill body. Each
mirrors the guard retired-artifact-cleanup.cts:77 already uses.

Severity stated accurately rather than dramatically: only the read at :755 needs
no race. _removeGsdEntries sweeps a pre-planted link at :1608 before the write
loop, and :880's stageDir is a fresh mkdtemp, so both of those require winning a
window. They are fixed as defense-in-depth, not as live exploits.

ALSO: the Changed changeset claimed "every command's observable behavior [is]
unchanged". Three rejection messages are reworded. It now says so, and says that
none of them reveals a host path it previously hid. A stale comment in
verify.cts still named validatePath; an init.cts warning hardcoded "resolves
outside the project directory" for a check that also rejects absolute paths, NUL
bytes and empty strings; and the rationale deleted with check-glossary-refs'
retired helper is restored, noting honestly that a rejected token is now
realpath-resolved before rejection rather than rejected by string comparison.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

scripts/check-glossary-refs.cjs

@@ -154,7 +154,11 @@ function isTracked(token) {
  * (CONTRIBUTING's "Do not compute a next number locally"), never a real path.
  */
 function extractTrackedRefs(text) {
-  const tokens = new Set();
+  // Maps token -> the ContainedPath tryWithinRoot returned for it. ADR-4650:
+  // the value that was validated for containment must be the exact value
+  // that gets probed later — never a path re-derived (e.g. re-joined) from
+  // the token, which could diverge from what was actually checked.
+  const tokens = new Map();
   const add = (raw) => {
     if (!PATH_TOKEN_RE.test(raw)) return;
     const token = raw.replace(/:\d+$/, '');
@@ -164,9 +168,18 @@ function extractTrackedRefs(text) {
     if (!/[A-Za-z0-9_]$/.test(token)) return;
     if (token.includes('NNNN')) return;
     if (!isTracked(token)) return;
-    // Containment decision is the canonical predicate's, per ADR-4650.
-    if (tryWithinRoot(token, ROOT, PathAcceptance.AbsoluteInsideRoot) === null) return;
-    tokens.add(token);
+    // Containment decision is the canonical predicate's, per ADR-4650. Carry
+    // the returned ContainedPath forward so checkFileRefs stats the SAME
+    // value that was validated, instead of re-joining `token` onto ROOT.
+    //
+    // The containment ANSWER is unchanged from the retired lexical-only
+    // `isWithinRoot`, but the canonical predicate resolves symlinks, so a
+    // rejected token is now realpath-resolved before being rejected rather
+    // than rejected by string comparison alone; the result is still never
+    // surfaced and the token is never stat'd unless it is contained.
+    const contained = tryWithinRoot(token, ROOT, PathAcceptance.AbsoluteInsideRoot);
+    if (contained === null) return;
+    tokens.set(token, contained);
   };
   const subTokenRe = /[\w.-]+(?:\/[\w.-]+)*/g;
   for (const line of text.split(/\r?\n/)) {
@@ -187,14 +200,17 @@ function extractTrackedRefs(text) {
 
 /** Check A: every tracked reference must resolve on disk. */
 function checkFileRefs(contextText) {
-  const tokens = [...extractTrackedRefs(contextText)].sort();
+  const entries = [...extractTrackedRefs(contextText)].sort(([a], [b]) => (a < b ? -1 : a > b ? 1 : 0));
   const findings = [];
-  for (const token of tokens) {
-    if (!fs.existsSync(path.join(ROOT, token))) {
+  for (const [token, contained] of entries) {
+    // Stat the ContainedPath returned by tryWithinRoot — NOT a re-joined
+    // path.join(ROOT, token) — so the path that was validated for
+    // containment is the path that is probed (ADR-4650).
+    if (!fs.existsSync(contained)) {
       findings.push(`CONTEXT.md references \`${token}\` which does not exist in the repo.`);
     }
   }
-  return { findings, checked: tokens.length };
+  return { findings, checked: entries.length };
 }
 
 /** The glossary's own claim: `Runtime enum: `allRuntimes` (N values: a, b, c)`. */