fix(#1264): harden unattended identity bootstrap

docs/reports/code-review/1264-code-review.md

@@ -1,59 +1,78 @@
-# Issue #1264 Independent Code and Security Review
+# Issue #1264 Code and Security Review
 
-> Scope: uncommitted delivery delta for `fix/1264-fleet-unattended-first-start` against
-> `origin/next@476db12b92971634b67fd2057b7577ee5894e449` | Reviewer: Codex CLI via Mosaic review tools
+> Branch: `fix/1264-fleet-unattended-first-start` | Base:
+> `origin/next@476db12b92971634b67fd2057b7577ee5894e449`
 
-## Initial code review
+## Initial automated review
 
-Command:
+Codex reviewed the pre-PR uncommitted delta with:
 
 ```bash
 ~/.config/mosaic/tools/codex/codex-code-review.sh --uncommitted \
   -o /tmp/1264-codex-code-review.json
 ```
 
-Result: `request-changes`, confidence `0.93`, `20` files reviewed, `0` blockers, `1` should-fix.
+Result: `request-changes`, confidence `0.93`, 20 files, one should-fix. `checkSoul()` trimmed
+`MOSAIC_AGENT_NAME` for pre-seed resolution while composition used the original value, so a padded
+name could seed files before later refusal.
 
-Finding: `checkSoul()` trimmed `MOSAIC_AGENT_NAME` for pre-seed roster resolution while later
-composition used the original value. A padded exact name could therefore seed identity files and
-then fail composition.
+Remediation rejected blank/leading/trailing-whitespace values before roster lookup or writes and
+added three built-CLI no-side-effect regressions. Automated re-review approved that delta with no
+findings (confidence `0.86`). Initial security review reported risk `none` (confidence `0.91`).
 
-Remediation:
+## Formal exact-head review
 
-- treat any present blank or surrounding-whitespace value as an invalid fleet launch;
-- reject it before roster lookup or identity writes; and
-- add three real-CLI no-side-effect regressions for leading padding, trailing padding, and empty
-  values.
+Daphne reviewed PR #1268 at exact head `43fa0477877e0d0f110da8d11c3033b40ddeb191` and filed Gitea
+review ID 168 as `REQUEST_CHANGES`. The review was source/PR-only; the canary remained untouched.
 
-## Code re-review
+Blocking groups:
 
-Command:
+1. class mismatch was validated after first-start mutation;
+2. secure `USER.md` validation was discarded before ordinary path-following composition;
+3. `existsSync()` treated a dangling destination symlink as missing, allowing counterpart partial
+   publication; and
+4. the built-CLI/evidence chain allowed stale ignored `dist/`, cited an unshipped canary object, and
+   carried conflicting test totals/pane wording.
+
+The diagnostic's defaults-only repair advice was also inaccurate for roster/class/destination
+failures.
+
+## Formal-review remediation
+
+All four blocking groups received regressions before production changes. The RED run produced four
+failures while 1,568 existing tests passed. Remediation then:
+
+- validates canonical name and class before seeding;
+- preflights destination directory entries with `lstatSync()` so target and dangling symlinks fail
+  before publication;
+- securely reads `USER.md` through an `O_NOFOLLOW` descriptor at composition time;
+- adds a Mosaic build before package Vitest and a dependency build in the clean-checkout command;
+- replaces defaults-only advice with neutral named-component repair guidance; and
+- reconciles shipping canary provenance, pane chronology, commands, and totals.
+
+Remediation code review:
 
 ```bash
 ~/.config/mosaic/tools/codex/codex-code-review.sh --uncommitted \
-  -o /tmp/1264-codex-code-rereview.json
+  -o /tmp/1264-remediation-code-review.json
 ```
 
-Result: `approve`, confidence `0.86`, `15` files reviewed, no findings. The review sandbox could run
-package typecheck but could not run Vitest because its checkout was read-only and Vite attempted to
-create a timestamped config artifact (`EROFS`). This is not scored as test evidence; the executor's
-writable worktree independently passed the focused and full suites recorded in the QA report.
+Result: `approve`, confidence `0.88`, 6 files, no findings. Summary: the fail-closed destination
+checks, class-validation order, secure composition, and build-before-Vitest path are coherent.
 
-## Security review
-
-Final command:
+Remediation security review:
 
 ```bash
 ~/.config/mosaic/tools/codex/codex-security-review.sh --uncommitted \
-  -o /tmp/1264-codex-security-rereview.json
+  -o /tmp/1264-remediation-security-review.json
 ```
 
-Result: risk `none`, confidence `0.91`, `20` files reviewed, `0` critical/high/medium/low findings.
-The review specifically confirmed roster validation before seeding, bounded no-symlink reads,
-no-clobber publication, unsafe/padded identity refusal, and fail-closed behavior before runtime.
+Result: risk `none`, confidence `0.93`, 9 files, no critical/high/medium/low findings. The sandbox
+could not run Vitest because Vite attempted to create a temporary config artifact on its read-only
+mount (`EROFS`); executor-owned focused and full results are recorded in the QA report.
 
-## Independent PR review gate
+## Remaining review gate
 
-Automated review is complete. The PR still requires a formal reviewer who is neither the implementation
-seat nor Fred, per the assignment. That review and CI status are recorded in the QA report when
-available.
+Daphne must re-review the next exact pushed head. This report cannot record that future verdict
+without changing the reviewed head, so the authoritative terminal verdict belongs to PR #1268's
+Gitea review record. Fred and goals are excluded as reviewers.