status(cfold): record fresh ghost probe

machine-docs/STATUS-cfold.md

@@ -58,8 +58,10 @@ Adversary verdict:
 Current work item:
 - full real-CI `!testme` sweep evidence is mostly assembled; one recipe (`ghost`) remains non-green for
   a cfold-neutral upgrade regression on the recipe/environment side
-- fresh artifact audit confirms historical build `185` passed on the same Ghost ref now failing in
-  probe build `559`, so no cfold-local code fix has been identified in this repo yet
+- fresh follow-up probes now show the Ghost upgrade failure is not confined to PR #4 / PR #5: a reopened
+  PR #3 at ref `720faa0b` also re-failed twice post-cfold (`568`, `569`) with the same shape
+- one fresh `!testme` comment on Ghost PR #3 also spawned multiple runs (`568`, `569`, `570`), so there
+  is likely a duplicate-trigger side issue to investigate separately from cfold itself
 
 ### M2 baseline matrix (built from live PR heads + fresh post-cfold evidence)
 
@@ -71,7 +73,7 @@ Current work item:
 | custom-html-tiny | PR #7 `526502ba` | 5 | 1 | build `510` -> L5 |
 | discourse | PR #2 `b7d8a244` | 5 | 3 | build `521` -> L5 |
 | drone | PR #1 `049438e1` | 5 | 1 | build `506` -> L5 |
-| ghost | PR #4 `d88f5801` | 5 | 4 | build `557` -> L1 (upgrade fail) |
+| ghost | PR #3 `720faa0b` | 5 | 4 | build `568` -> L1 (upgrade fail) |
 | hedgedoc | PR #1 `441c411c` | 5 | 2 | build `555` -> L5 |
 | immich | PR #2 `17f1649c` | 5 | 3 | build `522` -> L5 |
 | keycloak | PR #3 `bfe0d16f` | 5 | 3 | build `553` -> L5 |
@@ -90,17 +92,21 @@ Current work item:
 
 `ghost` is the only recipe still preventing an M2 claim.
 
-- Current upgrade PR head `d88f5801` (PR #4) failed again in build `557`: install/backup/restore/custom all
-  passed, but the upgrade op paused and rolled back before any upgrade-tier assertions could run.
-- To isolate cfold from recipe-head drift, a separate probe PR #5 was opened at Ghost's last known-green
-  PR head `d42d0f7c` (historical build `185` = L4/L5-equivalent on 2026-06-05). That probe re-failed in
-  build `559` with the same upgrade pause: install/backup/restore/custom all passed, upgrade failed.
-- Focused artifact audit after build `559` confirmed the same-ref comparison explicitly:
+- Current upgrade PR heads and fresh post-cfold outcomes are all red with the same stage shape:
+  - PR #3 `720faa0b`: builds `568` and `569` -> L1; install/backup/restore/custom/lint pass, upgrade fail
+  - PR #4 `d88f5801`: build `557` -> L1; install/backup/restore/custom pass, upgrade fail
+  - PR #5 `d42d0f7c`: build `559` -> L1; install/backup/restore/custom/lint pass, upgrade fail
+- Focused artifact audit still confirms the strongest same-ref comparison explicitly:
   historical build `185` (`d42d0f7c7cf9`) had `upgrade=pass`, while fresh build `559` on that same ref
   has `upgrade=fail` with the canonical `custom` stage still green.
+- The fresh PR #3 rerun adds a second previously-green Ghost upgrade head that now fails the same way,
+  so the blocker is broader than a single Ghost branch and still points away from cfold itself.
+- Side observation from the PR #3 retrigger: a single `!testme` comment at `2026-06-13T00:07:50Z` spawned
+  three new Ghost runs (`568`, `569`, `570`). `568` and `569` are already red with the same upgrade-only
+  failure; `570` was still in flight at the time of this status update.
 - Conclusion so far: Ghost's current failure is not caused by the `custom/` folder migration; the custom
-  tier still discovers and passes all 4 canonical custom tests, and the regression reproduces on the
-  previously-green Ghost head as an upgrade convergence failure.
+  tier still discovers and passes all 4 canonical custom tests, and the regression reproduces across
+  multiple Ghost PR heads as an upgrade convergence failure.
 
 ### Fresh Adversary state