diff --git a/k8s/bases/infrastructure/controllers/flux-operator/helm-release.yaml b/k8s/bases/infrastructure/controllers/flux-operator/helm-release.yaml
index 8b4320e3e..41deabeff 100644
--- a/k8s/bases/infrastructure/controllers/flux-operator/helm-release.yaml
+++ b/k8s/bases/infrastructure/controllers/flux-operator/helm-release.yaml
@@ -45,6 +45,40 @@ spec:
   #
   # Applied here rather than in the FluxInstance because the operator is
   # installed by Helm, so the FluxInstance's kustomize patches never reach it.
+
+  # C-0211 (CIS-5.7.3) — the two non-privilege fields, supplied at TEMPLATE level.
+  #
+  # flux-system is excluded from add-pod-security-context / add-container-security-context
+  # because its workloads need elevated privilege by design, and the namespace-scoped
+  # add-baseline-context-optin-* rules cannot close the gap here: all three match
+  # `kinds: Pod`, so they mutate at ADMISSION and leave the stored workload spec
+  # Kubescape reads untouched. Only a template-level change moves the measurement —
+  # the same reason #3541 (velero) and #3544 (the four FluxInstance controllers) were
+  # done this way rather than by labelling the namespace.
+  #
+  # Measured against the live PROD stored specs 2026-09-04: of flux-system's six
+  # workloads, four controllers already carry both fields from #3544, leaving
+  # flux-operator and tofu-controller. tofu-controller is retired and pending removal
+  # in #3480, so flux-operator is the namespace's last workload needing this.
+  #
+  # Both fields are inert here, which is the point rather than a caveat — this changes
+  # a declared posture without changing how the operator runs:
+  # - fsGroupChangePolicy governs the kubelet's volume ownership pass, which runs only
+  #   when a pod sets fsGroup. This pod sets none, and its only volumes are an emptyDir
+  #   and the projected service-account token — no PVC, no hostPath.
+  # - seLinuxOptions {level: s0} is the default non-privileged MCS level and a no-op on
+  #   this AppArmor cluster (runc ignores SELinux labels when SELinux is not the active
+  #   LSM), retained as CIS-5.7.3 defense-in-depth on any SELinux node.
+  #
+  # Neither carries the runAsGroup-without-runAsUser coupling documented above, so
+  # neither can stop the sandbox building.
+  #
+  # 🔴 The pod-level op adds a LEAF into the chart's pod securityContext, which renders
+  # as `{}` (read from the live Deployment 2026-09-04). A leaf add is deliberate: adding
+  # the whole object would REPLACE it and silently discard any field the chart starts
+  # setting later, which is exactly what the +(key) anchors elsewhere exist to prevent.
+  # The cost is that a chart that stopped rendering the object would fail this patch —
+  # loudly, on a Renovate chart bump with CI in front of it, never silently.
   postRenderers:
     - kustomize:
         patches:
@@ -61,3 +95,10 @@ spec:
               - op: add
                 path: /spec/template/spec/containers/0/securityContext/runAsGroup
                 value: 65532
+              - op: add
+                path: /spec/template/spec/containers/0/securityContext/seLinuxOptions
+                value:
+                  level: s0
+              - op: add
+                path: /spec/template/spec/securityContext/fsGroupChangePolicy
+                value: OnRootMismatch
diff --git a/scripts/validate-eks-ci-role-policy/main.go b/scripts/validate-eks-ci-role-policy/main.go
index 52bdf9d22..c8bdfe579 100644
--- a/scripts/validate-eks-ci-role-policy/main.go
+++ b/scripts/validate-eks-ci-role-policy/main.go
@@ -1655,7 +1655,39 @@ const (
 // The previous approved aggregate digest was:
 //
 //	686bf5ade0d7869f3a20d50a05a7ac083744f0e88ede4fc02298ab9f67dfdf19
-const expectedRenderedSurfaceSHA = "a16fbd577f8155e64aacc02cb624f26d9628d75658ce804215d9d69a3884bf92"
+//
+// Re-approved 2026-09-04 for the flux-operator C-0211 template fix (#3239), which
+// adds two securityContext ops to that HelmRelease's existing postRenderer patch.
+//
+// Conservation proven by rendering ALL FIVE overlays from both trees with the same
+// renderer and taking the set difference in BOTH directions over the complete
+// apiVersion|kind|namespace|name identity: 548 documents on main e37660b2 and 548 on
+// this branch — ZERO added, ZERO removed, ZERO renamed. Neither side carries a
+// duplicate identity, so that pairing is one-to-one.
+//
+// The ENTIRE content delta across all five roots is seven added lines inside one
+// document — the flux-operator HelmRelease — and nothing is removed anywhere:
+//
+//   - op: add   /spec/template/spec/containers/0/securityContext/seLinuxOptions  {level: s0}
+//   - op: add   /spec/template/spec/securityContext/fsGroupChangePolicy          OnRootMismatch
+//
+// Neither op is grant-bearing: no rule, verb, resource, group, subject,
+// ServiceAccount, role reference or policy moved on either side. The surface digest
+// moved only because that HelmRelease is itself a selected authorization-capable
+// document, so its text is covered by the aggregate — which is the tripwire working
+// as designed, not a privilege change.
+//
+// The value below is CI's own computed digest at head 5dc2d805, taken from the
+// failing run rather than recomputed locally: this validator pins the renderer to
+// kubectl v1.36.2 and refuses any other, and the host that prepared this change has
+// v1.36.1. The conservation proof above is renderer-robust because both sides were
+// rendered with the SAME local renderer; the digest is not, so it comes from the
+// pinned one.
+//
+// The previous approved aggregate digest was:
+//
+//	a16fbd577f8155e64aacc02cb624f26d9628d75658ce804215d9d69a3884bf92
+const expectedRenderedSurfaceSHA = "ab05bc2c95924e372c038f878872aa94e3bd81380daa96a283bd7f74e2132a69"
 
 // authorizationOverlayPaths lists every independently reconciled production
 // layer where an object can grant privileges to the aws/aws service account.