diff --git a/CHANGELOG.md b/CHANGELOG.md
index 6659f982..5bb5487b 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -16,6 +16,10 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 - Add `feature-flags` skill with a repo-agnostic base and a MetaMask Mobile overlay for version-gated remote flags. Marked `base: true` so it installs even when its domain is filtered out. ([#147](https://github.com/MetaMask/skills/pull/147))
 - Add `analytics` skill (`platform/analytics`, moved from `coding`) with a repo-agnostic base and a MetaMask Mobile overlay for the canonical tracking API. Marked `base: true` so it installs even when its domain is filtered out. ([#140](https://github.com/MetaMask/skills/pull/140))
 
+### Changed
+
+- Stop testing skills from prompting unit tests for static styling. Mobile and Extension layer policies now treat spacing, color, typography, and fixed layout props as GAP / ACCEPT, the unit references ban style-value assertions and style-only test IDs, and coding guidelines no longer require tests for every component edit.
+
 ## [0.3.1]
 
 ### Fixed
diff --git a/domains/coding/skills/coding-guidelines/repos/metamask-extension.md b/domains/coding/skills/coding-guidelines/repos/metamask-extension.md
index 508d4d13..fd91fe9e 100644
--- a/domains/coding/skills/coding-guidelines/repos/metamask-extension.md
+++ b/domains/coding/skills/coding-guidelines/repos/metamask-extension.md
@@ -586,7 +586,8 @@ export function formatBalance(
 
 ### Write Tests for All Components and Utilities
 
-- **ALWAYS write tests for new code**
+- **ALWAYS write tests for new behavior** (logic, state, and conditional UI)
+- **Do not** write tests for static styling changes (spacing, color, typography, class-name swaps); see `extension-testing` → `references/unit.md`
 - Tests reduce possibilities of errors and regressions
 - Ensure components behave as expected
 - Use Jest as the testing framework
@@ -779,7 +780,7 @@ Before submitting a PR, ensure:
 
 ### Testing
 
-- [ ] Unit tests written for components
+- [ ] Unit tests written for component behavior (not static styling)
 - [ ] Unit tests written for utilities
 - [ ] Tests follow naming conventions
 - [ ] Tests cover happy paths and error cases
diff --git a/domains/coding/skills/coding-guidelines/repos/metamask-mobile.md b/domains/coding/skills/coding-guidelines/repos/metamask-mobile.md
index e966062b..95f7425f 100644
--- a/domains/coding/skills/coding-guidelines/repos/metamask-mobile.md
+++ b/domains/coding/skills/coding-guidelines/repos/metamask-mobile.md
@@ -18,7 +18,7 @@ parent: coding-guidelines
 **Code Quality**:
 
 - TypeScript guidelines from contributor docs • Functional components + hooks • PascalCase (components) / camelCase (functions)
-- Reusable components/utilities • TSDoc format • Comprehensive tests following testing layers (below)
+- Reusable components/utilities • TSDoc format • Comprehensive tests following testing layers (below) • No tests for static styling changes (spacing, color, typography, fixed layout props)
 - Redux selectors: install **selector-patterns** (`yarn skills --include coding/selector-patterns --save`) when writing or updating them.
 
 **Testing layers** (Mobile — canonical policy: testing domain `knowledge/testing-layers.md`, installed beside `mobile-testing`):
diff --git a/domains/testing/knowledge/extension-testing-layers.md b/domains/testing/knowledge/extension-testing-layers.md
index 44d20300..3e947d85 100644
--- a/domains/testing/knowledge/extension-testing-layers.md
+++ b/domains/testing/knowledge/extension-testing-layers.md
@@ -57,6 +57,29 @@ Is this scenario worth covering? (distinct realistic regression)
                └─ No → GAP / ACCEPT (do not invent E2E)
 ```
 
+## Not worth covering: static presentation
+
+A change to static or default presentation is **GAP / ACCEPT** at every layer.
+Do not add or update a test for it. This includes:
+
+- Spacing, padding, margin, and alignment
+- Color, typography, and design-token values
+- Fixed truncation or other layout props that never change with state
+- Swapping one class name, SCSS rule, or style value for another
+
+Do not add a `data-testid` only so a test can read `style` or `className`, and
+do not add or refresh a snapshot because only class names or static styles
+changed. Verify these changes with screenshots (`visual-testing`) and the PR's
+before/after evidence.
+
+Presentation that depends on state or props is behavior. Test it, and assert
+the outcome a user or caller sees (text shown or hidden, disabled, ARIA state),
+not pixel or class values.
+
+Do not use a coverage delta as the reason to add or skip a test. Coverage can
+rise from a render that asserts nothing, and a real bug fix can leave coverage
+flat. Coverage reports only point at uncovered branches worth considering.
+
 ## Defaults
 
 | Layer           | Owns                                                                                                  | Existing homes                                                                                      |
diff --git a/domains/testing/knowledge/testing-layers.md b/domains/testing/knowledge/testing-layers.md
index 66d691e8..5fbf2774 100644
--- a/domains/testing/knowledge/testing-layers.md
+++ b/domains/testing/knowledge/testing-layers.md
@@ -49,6 +49,28 @@ Is this scenario worth covering? (distinct realistic regression)
             └─ No → GAP / ACCEPT (do not invent E2E)
 ```
 
+## Not worth covering: static presentation
+
+A change to static or default presentation is **GAP / ACCEPT** at every layer.
+Do not add or update a test for it. This includes:
+
+- Spacing, padding, margin, inset, and alignment
+- Color, typography, and design-token values
+- Unconditional `numberOfLines` / `ellipsizeMode` or other fixed layout props
+- Swapping one Tailwind class or style value for another
+
+Do not add a `testID` only so a test can read `style` or `className`. Verify
+these changes with screenshots (`mobile-visual-testing`) and the PR's
+before/after evidence.
+
+Presentation that depends on state or props is behavior. Test it, and assert
+the outcome a user or caller sees (text shown or hidden, disabled, accessibility
+state), not pixel or class values.
+
+Do not use a coverage delta as the reason to add or skip a test. Coverage can
+rise from a render that asserts nothing, and a real bug fix can leave coverage
+flat. Coverage reports only point at uncovered branches worth considering.
+
 ## Defaults
 
 | Layer              | Owns                                                                                               | File pattern                      |
diff --git a/domains/testing/skills/extension-testing/references/unit.md b/domains/testing/skills/extension-testing/references/unit.md
index 0d5fd4bd..1a60227d 100644
--- a/domains/testing/skills/extension-testing/references/unit.md
+++ b/domains/testing/skills/extension-testing/references/unit.md
@@ -9,6 +9,34 @@ Reference: [MetaMask Unit Testing Guidelines](https://github.com/MetaMask/contri
 - **ALWAYS use Jest** for unit tests (not Mocha or Tape)
 - Leverage Jest's built-in features: module mocks, timer mocks, snapshots, and parallel test execution
 
+### Do Not Test Static Presentation
+
+Choose the layer first via installed `knowledge/extension-testing-layers.md`. Static or default presentation is GAP / ACCEPT there, so a diff that only changes it needs no new or updated test.
+
+- **NEVER** assert spacing, color, typography, or fixed layout values (computed styles, `className` strings, design-token values)
+- **NEVER** add or refresh a snapshot because only class names, SCSS, or static styles changed
+- **NEVER** add a `data-testid` whose only use is reading `style` or `className`
+- **Do not** justify a test by coverage alone. A render that touches a new JSX line adds coverage without protecting behavior.
+- Verify static presentation with screenshots (`visual-testing`) and the PR's before/after evidence
+
+```typescript
+❌ WRONG:
+it('removes the extra padding from the price summary', () => {
+  render(<PriceSummary />);
+  expect(screen.getByTestId('price-summary')).not.toHaveClass('px-4');
+});
+```
+
+Presentation that depends on state or props is behavior. Test it by asserting what the user sees (text shown or hidden, disabled, ARIA state), not class names or style values.
+
+```typescript
+✅ CORRECT:
+it('disables the submit button when the amount is empty', () => {
+  render(<SendForm amount="" />);
+  expect(screen.getByRole('button', { name: 'Send' })).toBeDisabled();
+});
+```
+
 ### Test File Organization
 
 #### File Placement
@@ -188,6 +216,7 @@ async function withController(...args: WithControllerArgs) {
 - **ALWAYS use "render matches snapshot" or similar variants**
 - Add context when needed: "render matches snapshot when not enabled"
 - Remember: Snapshots only check for changes, NOT correctness
+- Do not add a snapshot to cover a static styling change; see [Do Not Test Static Presentation](#do-not-test-static-presentation)
 
 Examples:
 
@@ -748,6 +777,7 @@ Before submitting tests, ensure:
 ### Final Checks
 
 - [ ] Snapshot tests are named "render matches snapshot"
+- [ ] No tests or snapshot updates exist only to cover static styling
 - [ ] All tests pass and are deterministic
 - [ ] Tests run in isolation (can run individually)
 - [ ] No console errors or warnings
diff --git a/domains/testing/skills/mobile-testing/references/unit.md b/domains/testing/skills/mobile-testing/references/unit.md
index 6bb85c84..457c125b 100644
--- a/domains/testing/skills/mobile-testing/references/unit.md
+++ b/domains/testing/skills/mobile-testing/references/unit.md
@@ -21,6 +21,35 @@ Follow installed `knowledge/testing-layers.md` before choosing unit vs component
 - **Default for screen/view behavior:** `*.view.test.tsx` via [`component-view.md`](component-view.md) — not a broad RTL unit test that mocks hooks/selectors.
 - **This skill applies when:** pure helpers, local utilities, narrow component contracts, or the CV framework cannot cover the case yet (smallest focused unit test + note why).
 - **Do not** add new full-page `*.test.tsx` files that render a screen and mock Redux/hooks to force UI state; convert or write CV instead.
+- **Do not** add a unit test for static presentation (spacing, color, typography, fixed layout props). The layer policy marks it GAP / ACCEPT — see [Do Not Test Static Presentation](#do-not-test-static-presentation).
+
+## Do Not Test Static Presentation
+
+A diff that only changes static or default styling needs no new or updated test, even inside an existing `*.test.tsx`. These assertions do not catch regressions a user would notice, and they break whenever a design token changes.
+
+```tsx
+// ❌ WRONG - asserts padding values after a spacing fix
+const style = StyleSheet.flatten(
+  screen.getByTestId(SelectorsIDs.PRICE_SUMMARY).props.style,
+);
+expect(style.paddingLeft).toBeUndefined();
+
+// ❌ WRONG - asserts fixed truncation props; Jest never renders the overflow
+expect(title.props.numberOfLines).toBe(1);
+expect(title.props.ellipsizeMode).toBe('tail');
+
+// ❌ WRONG - asserts margins set by a Tailwind class
+expect(StyleSheet.flatten(divider.props.style)).toEqual(
+  expect.objectContaining({ marginBottom: 0, marginTop: 20 }),
+);
+```
+
+- **Do not** add a `testID` whose only use is reading `style`.
+- **Do not** remove a `useTailwind` mock so `StyleSheet.flatten` can see real spacing.
+- **Do not** justify a test by coverage alone. A render that touches a new JSX line adds coverage without protecting behavior.
+- Verify static presentation with screenshots (`mobile-visual-testing`) and the PR's before/after evidence.
+
+Presentation that depends on state or props is behavior. Test it by asserting what the user sees (text shown or hidden, disabled, accessibility state), not style values.
 
 ## Test Naming Rules
 
@@ -77,7 +106,7 @@ const createTestEvent = (overrides = {}) => ({
 
 - **ALWAYS prefer `testID` props** for selecting elements in tests
 - **Use `getByTestId`** as the primary query method for reliable element selection
-- **Add `testID` props** to components when writing new code or updating existing code
+- **Add `testID` props** when a justified behavior test needs a stable query — not just because a component was touched, and never only to read `style`
 - **Avoid selecting by text** when the text might change (i18n, copy updates)
 
 ```tsx
@@ -418,7 +447,9 @@ describe('MetaMetricsCustomTimestampPlugin', () => {
 
 ## Test Coverage (MANDATORY)
 
-**EVERY component MUST test:**
+Apply this list only after the layer policy says a unit test is justified. Cover the behavior branches the code actually has. A static styling change adds no branch, so it adds no test.
+
+**When a unit test is justified, it MUST cover:**
 
 - ✅ **Happy path** - normal expected behavior
 - ✅ **Edge cases** - null, undefined, empty values, boundary conditions
@@ -597,6 +628,7 @@ expect(result).toBe(false);
 Before submitting any test file, verify:
 
 - [ ] **No `toMatchSnapshot()` calls** — BANNED; use explicit assertions or `toMatchInlineSnapshot()` instead
+- [ ] **No static style assertions** — no `StyleSheet.flatten` padding/margin/color checks, fixed layout-prop checks, or style-only testIDs
 - [ ] **No mocking to inject testIDs** - Use component's built-in testID support
 - [ ] **testIDs via child prop objects** - Use `closeButtonProps={{ testID }}` not mocks
 - [ ] **No "should" in any test name**
@@ -617,6 +649,7 @@ Before submitting any test file, verify:
 
 - ❌ **Using `toMatchSnapshot()`** — BANNED; it writes opaque `.snap` files with no code owner; use explicit assertions or `toMatchInlineSnapshot()` instead
 - ❌ **Mocking to inject testIDs** - Components already support testID (see guidelines above)
+- ❌ **Testing static styling** - Spacing, color, and fixed layout props are verified visually, not in Jest
 - ❌ **Using "should" in test names** - This is the #1 mistake, use action-oriented descriptions
 - ❌ **Testing multiple behaviors in one test** - One test, one behavior
 - ❌ **Sharing state between tests** - Each test must be independent