diff --git a/STATUS.md b/STATUS.md
index 07a8ccc8..aff9354b 100644
--- a/STATUS.md
+++ b/STATUS.md
@@ -43,8 +43,13 @@ each with a number.
 
 **Rows.** One closed, five opened across the day. The register went from **264** to **269** open rows.
 
-**Needs you.** Nothing. Both new versions run on the demo HP box only; the fleet floor is unchanged at
-0.250.0. If you do nothing: the other boxes keep the version they have, and nothing breaks.
+**The floor is raised, as you asked.** The fleet minimum went from 0.250.0 to **0.253.0**. The N100
+demo box took it by itself in about twenty seconds and is healthy, with its other four apps still
+running; the HP demo box already had it. **The two boxes that are switched off did not get it** — Peti's
+has been off for 65 days on a much older version, and Tester 1 for a day. They will take it on their own
+whenever they come back, which is how a floor always works, and neither has been tried on this version.
+
+**Needs you.** Nothing. If you do nothing: the two sleeping boxes update themselves when they wake.
 
 ---
 
diff --git a/documentation/audits/i18n-slice2-2026-09-18/floor-raise-0.253.0.md b/documentation/audits/i18n-slice2-2026-09-18/floor-raise-0.253.0.md
new file mode 100644
index 00000000..70336ac2
--- /dev/null
+++ b/documentation/audits/i18n-slice2-2026-09-18/floor-raise-0.253.0.md
@@ -0,0 +1,48 @@
+# Fleet floor raised to 0.253.0 (2026-09-18, operator asked)
+
+**Before:** `min_controller_version` 0.250.0, `min_agent` 0.131.0.
+**After:** `min_controller_version` **0.253.0**, `min_agent` **0.131.0** (the value the v0.253.0
+CHANGELOG header declares; `minagent_header_gate.py` guarantees that line exists).
+
+The floor is ABOVE the vouched golden (0.246.0), so the declaration is what carries it — an
+undeclared floor above the golden is HELD by design (R-472, hub v0.112.0).
+
+## What it reached, and what it could not
+
+| customer | before | status | after |
+|---|---|---|---|
+| Demo Ügyfél (demo-felhom, N100) | 0.250.0 | ONLINE | **0.253.0**, in ~20 s |
+| Demo HP | 0.253.0 (hand-deployed for the live proof) | ONLINE | unchanged |
+| Peti Proxmox | 0.115.0 | **DOWN, last seen 65 days ago** | nothing — it takes the floor when it returns |
+| Tester 1 | 0.245.0 | **DOWN, last seen 1 day ago** | nothing — same |
+
+Agent versions checked first, because the floor is held for a box whose agent is below the
+requirement: demo-felhom **0.132.0**, demo-hp **0.132.0**, both above 0.131.0. (`drill-r50` is a dead
+drill host on 0.129.0 and reads `floor held` — correct, and it has no customer.)
+
+## Positive observables, both ends
+
+Hub:
+```
+[INFO] Global controller-version floor set to "0.253.0" (declared MinAgent "0.131.0")
+[INFO] managed floor SERVED for demo-felhom: floor 0.253.0, agent requirement "0.131.0"
+       from declared (golden 0.246.0)
+```
+The box (demo-felhom guest 9201), which is the half that matters — the hub SAYS, the box DOES:
+```
+/etc/felhom-controller-image = gitea.dooplex.hu/admin/felhom-controller:0.253.0
+docker ps                    = …:0.253.0  Up 26 seconds (healthy)
+[INFO] [offsite-apply] settle-gate: GO — at/above floor 0.253.0 (we are 0.253.0)
+```
+Its other four containers (traefik, cloudflared, filebrowser, opengist) stayed up across the update.
+
+And the release works on this SECOND box, in both languages — `/login?flash=flash.login.password_changed`:
+„Jelszó sikeresen módosítva. Kérjük, jelentkezzen be az új jelszóval." / "Your password is changed.
+Sign in with the new one."
+
+## What a reader should NOT conclude from this
+
+The two DOWN customers have not been tested on 0.253.0 and will take it unattended when they next
+report. Their boxes are on 0.115.0 and 0.245.0 — 0.115.0 in particular is a long way back, and no
+box has been carried from there to 0.253.0 in one step by anyone. That is the ordinary consequence of
+a floor and is not new today, but it is the honest reading of the table above.