diff --git a/.changeset/section-component-editability-boundary.md b/.changeset/section-component-editability-boundary.md
new file mode 100644
index 0000000000..da993b78ae
--- /dev/null
+++ b/.changeset/section-component-editability-boundary.md
@@ -0,0 +1,49 @@
+---
+"@objectstack/spec": patch
+---
+
+docs(spec): a form section / page component gates visibility only — say so, and tell `disabled` where it belongs (#7887)
+
+`FormSectionSchema` and `PageComponentSchema` declare no `disabled`, `readonly`
+or `readonlyWhen` slot. Writing one has always been a loud parse error, but a
+**bare** one: the message named the offending key and offered nothing, because
+there is nothing on those shapes to point a rename at. An author — increasingly
+an AI one — had no way to tell "this key is mis-spelled" from "this key belongs
+somewhere else entirely".
+
+It is the second. Ruled a **boundary, not a gap** (maintainer, 2026-08-12):
+sections and page components gate *visibility*; **editability lives on fields**.
+Neither shape has read-only semantics of its own for anything to enforce, so a
+slot here would be declared-but-unenforced from the day it landed — the ADR-0049
+class this repo is retiring elsewhere.
+
+So no key was added and no alias row was registered. What changed is the
+sentence the rejection carries. `disabled`, `disabledWhen`, `readonly`,
+`readOnly`, `readonlyWhen` and `editable` on either shape now answer with the
+boundary and the destination:
+
+> Editability is a FIELD-level concern. This shape gates VISIBILITY only — a
+> deliberate boundary, not a missing key (#7887): a section / page component has
+> no read-only semantics of its own to enforce. Write `readonly: true` (or the
+> conditional `readonlyWhen` predicate) on the form field(s) inside it instead;
+> to hide the whole section or component, use `visibleWhen`.
+
+It points at **`readonlyWhen`** and never at `disabledWhen`, which exists on no
+field surface: `field.zod.ts` renames `disabled` to `readonly` for exactly that
+reason.
+
+**Acceptance is unchanged, in both directions.** Every metadata document that
+parsed before parses identically, and every key rejected before is still
+rejected — a guidance string is not an accepted key, and the pins assert both.
+The package's public API surface does not move either: the new options table
+lives in `shared/editability-boundary.ts`, which the barrel deliberately does not
+re-export, alongside the `strictObject` machinery it belongs to.
+
+**The prescription is filed on those two shapes, not on the table they share.**
+`VISIBILITY_STRICT_OPTIONS` has a third consumer, `FormFieldSchema`, which is
+the one view/page shape that *does* answer `disabled` — through its own
+`disabled → readonly` rename. A guidance set consumes a key before the rename
+channel is ever consulted, so filing this family in the shared table would have
+replaced the family's one correct pointer with a redirect away from it. Section
+and component take a new `VISIBILITY_ONLY_STRICT_OPTIONS`; the field shape keeps
+the bare options and its message is byte-for-byte what it was.
diff --git a/packages/spec/src/shared/editability-boundary.test.ts b/packages/spec/src/shared/editability-boundary.test.ts
new file mode 100644
index 0000000000..a2f8327fbd
--- /dev/null
+++ b/packages/spec/src/shared/editability-boundary.test.ts
@@ -0,0 +1,256 @@
+// Copyright (c) 2026 ObjectStack. Licensed under the Apache-2.0 license.
+
+/**
+ * #7887 — the section / page-component **editability boundary**, asserted.
+ *
+ * The maintainer's ruling of 2026-08-12, operative sentence: *"`FormSectionSchema`
+ * / `PageComponentSchema` gate **visibility only**; editability lives on fields.
+ * No `disabled` / `readonly` / `disabledWhen` slot is added to those shapes, and
+ * no alias row is registered for them."* The deliverable is therefore text-face
+ * only: the accepted set does not move, the rejected set does not move, and the
+ * single thing that changes is the sentence an author reads when they write an
+ * editability key on a shape that has no editability semantics.
+ *
+ * ## What each section below would catch
+ *
+ * 1. **the prescription reaches a real author** — asserted on the actual
+ *    `unrecognized_keys` message from `safeParse`, never on the options table.
+ *    A row filed in a table nothing consults is exactly the dead-entry shape
+ *    `alias-integrity.test.ts` exists for; reading the message back is the only
+ *    assertion that cannot pass that way.
+ * 2. **it points at `readonlyWhen`, and never at `disabledWhen`** — `field.zod.ts`
+ *    renames `disabled → readonly` and records that "a field has `readonlyWhen`,
+ *    not `disabledWhen`" (#7832). A prescription naming `disabledWhen` would send
+ *    the author to a key that exists on no field surface, which is worse than the
+ *    bare rejection it replaced.
+ * 3. **the field surface is untouched** — the trap this card was tiered up for.
+ *    `VISIBILITY_STRICT_OPTIONS` is shared with `FormFieldSchema`, which answers
+ *    `disabled` through its OWN alias row. A guidanceSet match `continue`s past
+ *    the rename channel, so filing this family in the shared table would have
+ *    replaced the one correct pointer in the family with a redirect away from it.
+ * 4. **acceptance is byte-identical** — the lane's admission criterion. A
+ *    guidance string must never become an accepted key.
+ *
+ * The options table itself lives in `editability-boundary.ts`, which the
+ * `shared/index.ts` barrel deliberately does not re-export — so the package's
+ * public API surface does not move either (`check:api-surface`), which is the
+ * same claim one level out.
+ */
+
+import { describe, it, expect } from 'vitest';
+
+import { FormFieldSchema, FormSectionSchema } from '../ui/view.zod';
+import { PageComponentSchema } from '../ui/page.zod';
+import { VISIBILITY_ONLY_STRICT_OPTIONS } from './editability-boundary';
+import { keySetMatches } from './suggestions.zod';
+
+/**
+ * The `unrecognized_keys` message for `value`, or a loud failure.
+ *
+ * Same helper, same reasoning, as `visible-when-alias-guidance.test.ts`: the
+ * probe bodies are minimal and may also miss a required key, so a whole-error
+ * stringify could let an assertion pass on text from an unrelated issue.
+ */
+function unknownKeyMessage(
+  schema: { safeParse: (v: unknown) => { success: boolean; error?: unknown } },
+  value: unknown,
+): string {
+  const r = schema.safeParse(value);
+  expect(r.success, `expected REJECTION, got a successful parse of ${JSON.stringify(value)}`).toBe(false);
+  const issues = (r.error as { issues?: Array<{ code?: string; message?: string }> }).issues ?? [];
+  const hit = issues.find((i) => i.code === 'unrecognized_keys');
+  expect(hit, `no \`unrecognized_keys\` issue in ${JSON.stringify(issues)}`).toBeDefined();
+  return hit?.message ?? '';
+}
+
+/** Minimal bodies that reach each surface's unknown-key path. */
+const SECTION = { fields: [] } as const;
+const COMPONENT = { type: 'text' } as const;
+const FORM_FIELD = { field: 'probe' } as const;
+
+/** The two shapes the ruling names, and nothing else. */
+const VISIBILITY_ONLY: ReadonlyArray<[string, { safeParse: (v: unknown) => { success: boolean; error?: unknown } }, object]> = [
+  ['FormSectionSchema', FormSectionSchema, SECTION],
+  ['PageComponentSchema', PageComponentSchema, COMPONENT],
+];
+
+/** The spellings the boundary set answers. */
+const EDITABILITY_KEYS = ['disabled', 'disabledWhen', 'readonly', 'readOnly', 'readonlyWhen', 'editable'] as const;
+
+// ===========================================================================
+// 1. The guidance reaches an author — on the real parse error
+// ===========================================================================
+describe('#7887 — the boundary prescription an author actually sees', () => {
+  it.each(VISIBILITY_ONLY)('%s answers `disabled` with the boundary, not a bare refusal', (_n, schema, base) => {
+    const m = unknownKeyMessage(schema, { ...base, disabled: true });
+    expect(m).toContain('Editability is a FIELD-level concern');
+    expect(m).toContain('gates VISIBILITY only');
+    // Rendered through the shared template's prescription channel — the bullet
+    // is what puts it directly after the key statement and before the history
+    // sentence (#5955's ordering, pinned for this family in `ui/view.test.ts`).
+    expect(m).toContain('\n  • Editability is a FIELD-level concern');
+  });
+
+  it.each(VISIBILITY_ONLY)('%s answers the whole editability family, not just `disabled`', (_n, schema, base) => {
+    for (const key of EDITABILITY_KEYS) {
+      expect(
+        unknownKeyMessage(schema, { ...base, [key]: 'x' }),
+        `\`${key}\` should reach the boundary prescription`,
+      ).toContain('Editability is a FIELD-level concern');
+    }
+  });
+
+  it.each(VISIBILITY_ONLY)('%s emits the prescription ONCE for a body carrying several of them', (_n, schema, base) => {
+    // The property that makes this a SET rather than N exact entries: one
+    // paragraph per message, however many members were written.
+    const m = unknownKeyMessage(schema, { ...base, disabled: true, readonly: true, editable: false });
+    expect(m.split('Editability is a FIELD-level concern')).toHaveLength(2);
+    // …and every offending key is still named.
+    for (const key of ['disabled', 'readonly', 'editable']) expect(m).toContain(`\`${key}\``);
+  });
+
+  it.each(VISIBILITY_ONLY)('%s still puts the history sentence last (the #5955 order survives the new set)', (_n, schema, base) => {
+    const m = unknownKeyMessage(schema, { ...base, disabled: true });
+    const history = 'Before ADR-0089 D3a these were dropped silently';
+    expect(m.indexOf('Editability is a FIELD-level concern')).toBeLessThan(m.indexOf(history));
+  });
+});
+
+// ===========================================================================
+// 2. It names `readonlyWhen` — and must never name `disabledWhen`
+// ===========================================================================
+describe('#7887 — the prescription points at a key that exists', () => {
+  it.each(VISIBILITY_ONLY)('%s names the field-level `readonly` / `readonlyWhen` pair', (_n, schema, base) => {
+    const m = unknownKeyMessage(schema, { ...base, disabled: true });
+    expect(m).toContain('`readonly: true`');
+    expect(m).toContain('`readonlyWhen`');
+  });
+
+  it.each(VISIBILITY_ONLY)('%s never names `disabledWhen` — no field surface declares it', (_n, schema, base) => {
+    // `field.zod.ts` renames `disabled → readonly` precisely because a field
+    // has `readonlyWhen`, not `disabledWhen` (#7832). Pointing at the latter
+    // would be a rejection that hands the author their next rejection.
+    const m = unknownKeyMessage(schema, { ...base, disabledWhen: 'record.locked' });
+    expect(m).toContain('Editability is a FIELD-level concern');
+    // The offending key is echoed back in the front matter, so the prohibition
+    // is on the PRESCRIPTION text, which is everything after the bullet.
+    const prescription = m.slice(m.indexOf('\n  • '));
+    expect(prescription).not.toContain('disabledWhen');
+  });
+
+  it('the boundary also names the visibility escape hatch, and that key really is accepted', () => {
+    const m = unknownKeyMessage(FormSectionSchema, { ...SECTION, disabled: true });
+    expect(m).toContain('`visibleWhen`');
+    expect(FormSectionSchema.safeParse({ ...SECTION, visibleWhen: 'record.x' }).success).toBe(true);
+    expect(PageComponentSchema.safeParse({ ...COMPONENT, visibleWhen: 'record.x' }).success).toBe(true);
+  });
+});
+
+// ===========================================================================
+// 3. The field surface is UNCHANGED — the shared-table trap
+// ===========================================================================
+describe('#7887 — `FormFieldSchema` sees exactly what it saw before', () => {
+  it('`disabled` on a form field still renames onto `readonly`, with no boundary text', () => {
+    const m = unknownKeyMessage(FormFieldSchema, { ...FORM_FIELD, disabled: true });
+    expect(m).toContain('Did you mean `disabled` → `readonly`?');
+    // The regression this whole file exists to catch. Hoisting
+    // `EDITABILITY_BOUNDARY_KEYS` into `VISIBILITY_STRICT_OPTIONS` makes the set
+    // fire here, and a set match `continue`s past the rename channel — so the
+    // line above would vanish and this line would appear, redirecting a field
+    // author AWAY from the one surface where `readonly` is real.
+    expect(m).not.toContain('Editability is a FIELD-level concern');
+  });
+
+  it('the boundary options are filed on the two visibility-only shapes, never the shared table', () => {
+    const names = (VISIBILITY_ONLY_STRICT_OPTIONS.guidanceSets ?? []).map((s) => s.name);
+    expect(names).toContain('EDITABILITY_BOUNDARY_KEYS');
+    // The ADR-0089 set is still there and still first — the boundary set is an
+    // addition, not a replacement.
+    expect(names[0]).toBe('VISIBILITY_KEY_PATTERN');
+  });
+
+  it('no editability key matches `VISIBILITY_KEY_PATTERN`, so set order is not load-bearing', () => {
+    // Both sets live on one table. If a future edit widened the visibility
+    // pattern to reach (say) `editable`, declaration order would silently start
+    // deciding which prescription an author reads — the tie-break
+    // `alias-integrity.test.ts` forbids any in-repo table from depending on.
+    const visibility = (VISIBILITY_ONLY_STRICT_OPTIONS.guidanceSets ?? [])
+      .find((s) => s.name === 'VISIBILITY_KEY_PATTERN');
+    expect(visibility).toBeDefined();
+    for (const key of EDITABILITY_KEYS) {
+      expect(keySetMatches(visibility!, key), `\`${key}\` is claimed by both sets`).toBe(false);
+    }
+  });
+
+  it('no alias row was registered for the boundary — the ruling forbids one', () => {
+    // "no alias row is registered for them — an alias would declare a key the
+    // runtime does not honour". An alias TARGET must be a key the shape accepts
+    // (`alias-integrity.test.ts`), and neither shape accepts any of these, so a
+    // row here would be a pointer into a second rejection.
+    for (const table of [VISIBILITY_ONLY_STRICT_OPTIONS]) {
+      for (const key of EDITABILITY_KEYS) {
+        expect(table.aliases?.[key], `\`${key}\` must not have an alias row`).toBeUndefined();
+      }
+    }
+  });
+});
+
+// ===========================================================================
+// 4. Acceptance is byte-identical — a guidance string is not a key
+// ===========================================================================
+describe('#7887 — no acceptance change', () => {
+  it.each(VISIBILITY_ONLY)('%s still REJECTS every editability spelling', (_n, schema, base) => {
+    for (const key of EDITABILITY_KEYS) {
+      expect(
+        schema.safeParse({ ...base, [key]: true }).success,
+        `\`${key}\` must stay rejected — this card curates messages, it does not widen the shape`,
+      ).toBe(false);
+    }
+  });
+
+  it('a representative section that parsed before still parses, unchanged in output', () => {
+    const authored = {
+      name: 'billing',
+      label: 'Billing',
+      description: 'Invoicing details',
+      collapsible: true,
+      collapsed: false,
+      columns: 2,
+      visibleOn: 'record.type == "customer"',
+      fields: ['amount', { field: 'currency', readonly: true }],
+    };
+    const r = FormSectionSchema.safeParse(authored);
+    expect(r.success).toBe(true);
+    // The ADR-0089 fold still runs, and the field-level `readonly` inside is
+    // still the accepted way to say what `disabled` on the section cannot.
+    // `ExpressionInputSchema` normalizes the authored string to a
+    // `{ dialect, source }` pair; the fold is about WHICH KEY carries it.
+    expect((r.data as { visibleWhen?: { source?: string } }).visibleWhen?.source)
+      .toBe('record.type == "customer"');
+    expect((r.data as { visibleOn?: unknown }).visibleOn).toBeUndefined();
+  });
+
+  it('a representative page component that parsed before still parses, unchanged in output', () => {
+    const authored = {
+      type: 'record:form',
+      id: 'main_form',
+      label: 'Details',
+      properties: { columns: 2 },
+      className: 'p-4',
+      visibility: 'current_user.is_admin',
+    };
+    const r = PageComponentSchema.safeParse(authored);
+    expect(r.success).toBe(true);
+    expect((r.data as { visibleWhen?: { source?: string } }).visibleWhen?.source)
+      .toBe('current_user.is_admin');
+    expect((r.data as { visibility?: unknown }).visibility).toBeUndefined();
+  });
+
+  it.each(VISIBILITY_ONLY)('%s keeps its bare message for a key in no family at all', (_n, schema, base) => {
+    // The new set must claim the editability family and nothing beyond it: an
+    // unrelated typo still gets the front matter plus history and no bullet.
+    const m = unknownKeyMessage(schema, { ...base, totallyUnrelatedKey: true });
+    expect(m).toContain('`totallyUnrelatedKey`');
+    expect(m).not.toContain('  • ');
+  });
+});
diff --git a/packages/spec/src/shared/editability-boundary.ts b/packages/spec/src/shared/editability-boundary.ts
new file mode 100644
index 0000000000..3f7229aa39
--- /dev/null
+++ b/packages/spec/src/shared/editability-boundary.ts
@@ -0,0 +1,109 @@
+// Copyright (c) 2026 ObjectStack. Licensed under the Apache-2.0 license.
+
+/**
+ * # The section / page-component **editability boundary** (#7887)
+ *
+ * Maintainer ruling, 2026-08-12: `FormSectionSchema` (`ui/view.zod.ts`) and
+ * `PageComponentSchema` (`ui/page.zod.ts`) gate **visibility only**; editability
+ * lives on fields. No `disabled` / `readonly` / `readonlyWhen` slot is added to
+ * either shape, and no alias row is registered for them — an alias names a key
+ * the shape must then accept, and one the runtime does not honour is the
+ * ADR-0049 declared-but-unenforced class this repo is retiring elsewhere.
+ *
+ * What the ruling *does* buy the author is this module: the rejection stops
+ * being bare and starts naming where the key belongs.
+ *
+ * ## Package-internal on purpose — this module is NOT in `shared/index.ts`
+ *
+ * It sits beside `strict-object.ts` and `alias-probe.ts` in the set of shared
+ * modules the barrel deliberately does not re-export. An unknown-key options
+ * table is machinery for declaring schemas in *this* package, and its own type
+ * (`StrictObjectOptions`) is not public either — a published const of an
+ * unpublishable type is an export no consumer can even annotate.
+ *
+ * It also keeps the #7887 claim exactly true: the card's whole deliverable is
+ * that nothing observable moves except the sentence an author reads, and the
+ * package's public API surface (`check:api-surface`) does not move at all.
+ */
+
+import type { StrictObjectOptions } from './strict-object';
+import type { KeySetGuidance } from './suggestions.zod';
+import { VISIBILITY_STRICT_OPTIONS } from './visibility';
+
+/**
+ * The editability vocabulary an author reaches for on a shape that gates
+ * **visibility only**.
+ *
+ * Every spelling here is rejected by `FormSectionSchema` and
+ * `PageComponentSchema` today and stays rejected: this set changes the MESSAGE,
+ * never the verdict. `readOnly` sits alongside `readonly` because set
+ * membership is matched case-sensitively (the rename channel is what folds
+ * case, and a set match `continue`s past it).
+ */
+const EDITABILITY_BOUNDARY_KEYS = [
+  'disabled',
+  'disabledWhen',
+  'readonly',
+  'readOnly',
+  'readonlyWhen',
+  'editable',
+] as const;
+
+/**
+ * The ruling rendered as the thing an author actually reads: **boundary, not
+ * gap.**
+ *
+ * Deliberately points at **`readonlyWhen`** and not at `disabledWhen`:
+ * `field.zod.ts` renames `disabled → readonly` and records in its own comment
+ * that "a field has `readonlyWhen`, not `disabledWhen`" (#7832). Naming
+ * `disabledWhen` here would send an author to a key that exists on no field
+ * surface at all — a rejection that hands them their next rejection.
+ */
+const EDITABILITY_BOUNDARY_GUIDANCE: KeySetGuidance = {
+  name: 'EDITABILITY_BOUNDARY_KEYS',
+  keys: EDITABILITY_BOUNDARY_KEYS,
+  prescription:
+    'Editability is a FIELD-level concern. This shape gates VISIBILITY only — a '
+    + 'deliberate boundary, not a missing key (#7887): a section / page component has '
+    + 'no read-only semantics of its own to enforce. Write `readonly: true` (or the '
+    + 'conditional `readonlyWhen` predicate) on the form field(s) inside it instead; to '
+    + 'hide the whole section or component, use `visibleWhen`.',
+};
+
+/**
+ * {@link VISIBILITY_STRICT_OPTIONS} for the two shapes that gate visibility and
+ * **nothing else** — `FormSectionSchema` and `PageComponentSchema`.
+ *
+ * ## Why the boundary prescription is filed HERE and not in the shared table
+ *
+ * `VISIBILITY_STRICT_OPTIONS` has **three** consumers, and the third —
+ * `FormFieldSchema` — is the one view/page shape that *does* answer `disabled`,
+ * through its own `aliases: { disabled: 'readonly' }` row (`view.zod.ts`, whose
+ * comment at that site rejects shared-table filing for exactly this reason).
+ * Adding `EDITABILITY_BOUNDARY_KEYS` to the shared options would land it on that
+ * table too, and the consequences are not cosmetic:
+ *
+ * - `strictUnknownKeyError` consults exact `guidance` → `guidanceSets` →
+ *   `aliases`, and a set match `continue`s past the rename. The field author who
+ *   writes `disabled` would stop seeing *"Did you mean `disabled` → `readonly`?"*
+ *   and start being told editability is somewhere else — on the one surface
+ *   where it is right there.
+ * - `alias-integrity.test.ts` would go **red**, not quietly wrong, in two
+ *   places: its #7889 check fails any alias row a guidanceSet on the same table
+ *   already consumes, and its #6619 check fails a set member the shape
+ *   *declares* — which `readonly` is, on `FormFieldSchema`.
+ *
+ * So the two visibility-only shapes take these options and `FormFieldSchema`
+ * keeps the bare ones. The prescription text is written once, here.
+ */
+export const VISIBILITY_ONLY_STRICT_OPTIONS: StrictObjectOptions = {
+  ...VISIBILITY_STRICT_OPTIONS,
+  guidanceSets: [
+    // Declaration order decides among sets. Nothing in
+    // `EDITABILITY_BOUNDARY_KEYS` matches `VISIBILITY_KEY_PATTERN`
+    // (`/vis|conceal|hidden|show.?when/i`), so the order is not load-bearing —
+    // pinned in `editability-boundary.test.ts` so it cannot quietly become so.
+    ...(VISIBILITY_STRICT_OPTIONS.guidanceSets ?? []),
+    EDITABILITY_BOUNDARY_GUIDANCE,
+  ],
+};
diff --git a/packages/spec/src/shared/visible-when-alias-guidance.test.ts b/packages/spec/src/shared/visible-when-alias-guidance.test.ts
index 1b71027cfd..6bd493e224 100644
--- a/packages/spec/src/shared/visible-when-alias-guidance.test.ts
+++ b/packages/spec/src/shared/visible-when-alias-guidance.test.ts
@@ -174,10 +174,17 @@ describe('#7832 — the deliberate gaps (an alias here would name a key the shap
     ['SelectOptionSchema', SelectOptionSchema, OPTION],
     ['FormSectionSchema', FormSectionSchema, SECTION],
     ['PageComponentSchema', PageComponentSchema, COMPONENT],
-  ])('%s declares no `disabledWhen` / `disabled` / `readonly`, so `disabled` stays uncurated', (_n, schema, base) => {
-    // Rejected — loudly, with the surface named — just without a pointer,
-    // because there is nothing on this shape to point at. If any of these ever
+  ])('%s declares no `disabledWhen` / `disabled` / `readonly`, so no alias row is filed', (_n, schema, base) => {
+    // Rejected — loudly, with the surface named — and with no RENAME, because
+    // there is nothing on this shape to point a rename at. If any of these ever
     // gains a disabled-ish key, this assertion fails and the row becomes owed.
+    //
+    // #7887 ruled the absence a BOUNDARY rather than a gap and gave the two
+    // view/page shapes a prescription saying so (`EDITABILITY_BOUNDARY_KEYS`,
+    // pinned in `editability-boundary.test.ts`). That is the guidance channel,
+    // not the alias channel: no key was added, no alias row was registered, and
+    // every assertion below is unchanged and green by construction.
+    // `SelectOptionSchema` was out of that ruling's scope and stays bare.
     for (const target of ['disabledWhen', 'disabled', 'readonly']) {
       expect(
         schema.safeParse({ ...base, [target]: 'x' }).success,
diff --git a/packages/spec/src/ui/page.zod.ts b/packages/spec/src/ui/page.zod.ts
index 95ca7186cd..d280c255f5 100644
--- a/packages/spec/src/ui/page.zod.ts
+++ b/packages/spec/src/ui/page.zod.ts
@@ -3,7 +3,8 @@
 import { z } from 'zod';
 import { SnakeCaseIdentifierSchema } from '../shared/identifiers.zod';
 import { ExpressionInputSchema } from '../shared/expression.zod';
-import { normalizeVisibleWhen, VISIBILITY_STRICT_OPTIONS } from '../shared/visibility';
+import { normalizeVisibleWhen } from '../shared/visibility';
+import { VISIBILITY_ONLY_STRICT_OPTIONS } from '../shared/editability-boundary';
 import { SortItemSchema } from '../shared/enums.zod';
 import { FilterConditionSchema } from '../data/filter.zod';
 import { I18nLabelSchema, AriaPropsSchema } from './i18n.zod';
@@ -100,8 +101,26 @@ export const ElementDataSourceSchema = lazySchema(() => strictObject({
  * now under `alias-integrity.test.ts` and with an edit-distance rename for the
  * page-component keys the hand-written map had no channel for (`classNam` →
  * `className`).
+ *
+ * ## A component gates VISIBILITY, not editability (#7887 — boundary, not gap)
+ *
+ * There is no `disabled`, `readonly` or `readonlyWhen` on a page component, and
+ * that is a **deliberate boundary** ruled on 2026-08-12, not a slot nobody got
+ * round to adding: **editability lives on fields.** A component decides whether
+ * it renders at all (`visibleWhen`); whether an input inside it can be edited is
+ * the field's own `readonly` / `readonlyWhen`, enforced by the field renderer
+ * that owns the input. Nothing in the platform reads a component-level read-only
+ * flag, so declaring one would ship the ADR-0049 declared-but-unenforced shape
+ * this repo is retiring elsewhere.
+ *
+ * Writing one anyway stays a loud parse error — unchanged — and since #7887 that
+ * error carries `EDITABILITY_BOUNDARY_KEYS`' prescription naming the field-level
+ * keys, so the author is redirected instead of merely refused. A widget with its
+ * own enabled/disabled notion expresses it inside {@link
+ * PageComponentSchema.properties}, which is that widget's own contract
+ * (`component.zod.ts`) and not this shape's.
  */
-export const PageComponentSchema = lazySchema(() => strictObject(VISIBILITY_STRICT_OPTIONS, {
+export const PageComponentSchema = lazySchema(() => strictObject(VISIBILITY_ONLY_STRICT_OPTIONS, {
   /** Definition */
   type: z.union([
     PageComponentType,
diff --git a/packages/spec/src/ui/view.zod.ts b/packages/spec/src/ui/view.zod.ts
index 8c2a67d5ce..1bc31cdfe8 100644
--- a/packages/spec/src/ui/view.zod.ts
+++ b/packages/spec/src/ui/view.zod.ts
@@ -7,6 +7,7 @@ import { strictObject, strictObjectError } from '../shared/strict-object';
 import { SnakeCaseIdentifierSchema } from '../shared/identifiers.zod';
 import { ExpressionInputSchema } from '../shared/expression.zod';
 import { normalizeVisibleWhen, VISIBILITY_STRICT_OPTIONS } from '../shared/visibility';
+import { VISIBILITY_ONLY_STRICT_OPTIONS } from '../shared/editability-boundary';
 import { I18nLabelSchema, AriaPropsSchema } from './i18n.zod';
 import { ChartTypeSchema } from './chart.zod';
 import { SharingConfigSchema } from './sharing.zod';
@@ -1772,6 +1773,16 @@ const FormFieldBaseSchema = lazySchema(() => {
       // match `VISIBILITY_KEY_PATTERN` and are already answered by the shared
       // ADR-0089 prescription, which consumes the key before the rename channel
       // is consulted, so an alias for either would be dead on arrival.
+      //
+      // #7887 filed the OTHER half of that split from the same reasoning and in
+      // the same direction: the two sibling shapes now carry an editability
+      // BOUNDARY prescription (`shared/editability-boundary.ts`), and it is
+      // filed on those two rather than shared, because a `disabled`-matching
+      // guidanceSet on THIS table would consume the key before the rename below
+      // ever runs — killing the one pointer that is correct here, and turning
+      // `alias-integrity.test.ts`'s #7889 reachability check red. The two
+      // decisions are one rule read from both ends: the shared table may only
+      // carry what is true of all three surfaces.
       aliases: { disabled: 'readonly' },
     }, shape),
   });
@@ -1847,8 +1858,26 @@ export const FormFieldSchema: z.ZodType<FormField, FormFieldInput> = lazySchema(
  * set-keyed form and the conversion became the same one call every other closed
  * shape in this package makes. The `.transform()` that folds
  * `visibleOn` → `visibleWhen` is unchanged and still runs after the parse.
- */
-export const FormSectionSchema = lazySchema(() => strictObject(VISIBILITY_STRICT_OPTIONS, {
+ *
+ * ## A section gates VISIBILITY, not editability (#7887 — boundary, not gap)
+ *
+ * There is no `disabled`, `readonly` or `readonlyWhen` on a section, and that is
+ * a **deliberate boundary** ruled on 2026-08-12, not a slot nobody got round to
+ * adding: **editability lives on fields.** A section decides whether its fields
+ * are *shown* (`visibleWhen`, plus `collapsible` / `collapsed` for how); whether
+ * a shown field can be *edited* is the field's own `readonly` /
+ * {@link FormFieldSchema.readonlyWhen}, enforced by the field renderer that owns
+ * the input. Nothing in the platform reads a section-level read-only flag, so
+ * declaring one would ship the ADR-0049 declared-but-unenforced shape this repo
+ * is retiring elsewhere.
+ *
+ * Writing one anyway stays a loud parse error — unchanged — and since #7887 that
+ * error carries `EDITABILITY_BOUNDARY_KEYS`' prescription naming the field-level
+ * keys, so the author is redirected instead of merely refused. To make a whole
+ * section non-editable, mark its fields `readonly`; to make it conditionally
+ * non-editable, give each field a `readonlyWhen` predicate.
+ */
+export const FormSectionSchema = lazySchema(() => strictObject(VISIBILITY_ONLY_STRICT_OPTIONS, {
   /**
    * Stable identifier for translation lookup. snake_case convention.
    * When provided, translation bundles can target this section's `label`