diff --git a/workplans/CUST-WP-0068-cache-only-work-record-recovery.md b/workplans/CUST-WP-0068-cache-only-work-record-recovery.md
index 3fcb4c3..84181c7 100644
--- a/workplans/CUST-WP-0068-cache-only-work-record-recovery.md
+++ b/workplans/CUST-WP-0068-cache-only-work-record-recovery.md
@@ -191,26 +191,49 @@ that disposition beats migration.
 Acceptance: a ruling recorded as a decision, with the chosen path executed or
 explicitly deferred.
 
-## Disposition the 23 orphans
+## Repair the binding gap before counting any orphans
 
 ```task
 id: CUST-WP-0068-T07
 status: todo
-priority: medium
+priority: high
 state_hub_task_id: "f44537f2-a738-5f39-b4c1-3711de06632e"
 ```
 
-Twenty-three records have no backing file, spread thinly across sixteen
-repositories — `ground-game` and `reuse-surface` (3 each), `rapp-core-hub`,
-`binky-control` and `railiance-fabric` (2 each), then one apiece.
+**The orphan count is not trustworthy and must be re-measured first.**
 
-`ADR-010`'s orphan rules apply: broken links where a file exists but was never
-recorded are metadata repairs; open hub-first records are `ADR-001` violations
-needing a file written or an explicit drop; closed hub-first records may be
-retained as provenance where cheap. Do not preserve open records that nothing
-can re-derive.
+`backing_filename` is written only by `PUT /workplans/index/bindings`, which
+`fix-consistency` calls for workplans that already carry a UUID. The documented
+registration flow mints the UUID *after* that sync — `fix-consistency` runs,
+C-06 skips because the workstation is not the registrar, then
+`registrar-reconcile` mints — so a newly registered workplan is left with no
+backing file recorded. Reproduced on `CUST-WP-0068` itself: it registered with
+`backing_filename: None`, and a second `fix-consistency` run populated it.
 
-Acceptance: every orphan is repaired, given a file, or dropped in writing.
+Consequence: **278 of 800 workplans on central (35%) record no backing file**,
+including 8 `proposed`, 4 `ready` and 4 `active`. By the read model's own
+definition those are hub-only records, and `ADR-010` predicted exactly this as
+"broken links — a file exists but `backing_filename` was never recorded …
+likely the largest class". The 23 cache-only records classified as orphans were
+counted with the same unreliable field, so that number is a starting point for
+investigation, not a work list.
+
+Order of work:
+
+1. Fix the ordering so registration binds in one pass — either the registrar
+   syncs bindings after minting, or the session protocol requires the second
+   `fix-consistency` run explicitly. The former is preferable; a protocol step
+   everyone must remember is how this gap opened.
+2. Re-sync bindings fleet-wide and re-measure. Also explain
+   `bindings: synced 63/73` — ten workplans did not bind and the reason is
+   unknown.
+3. Only then apply `ADR-010`'s orphan rules to whatever genuinely has no file:
+   open hub-first records are `ADR-001` violations needing a file written or an
+   explicit drop; closed ones may be retained as provenance where cheap.
+
+Acceptance: registration records a backing file in one pass; the fleet-wide
+count of file-less workplans is re-measured and explained; every remaining true
+orphan is repaired, given a file, or dropped in writing.
 
 ## Converge and retire the cache database