diff --git a/apps/frontend/src/stores/__tests__/engines.store.test.ts b/apps/frontend/src/stores/__tests__/engines.store.test.ts
index ba9835a..d7d8756 100644
--- a/apps/frontend/src/stores/__tests__/engines.store.test.ts
+++ b/apps/frontend/src/stores/__tests__/engines.store.test.ts
@@ -1,59 +1,74 @@
 import { describe, it, expect } from 'vitest'
 import { getEngineDigest } from '../engines.store'
 
-const WIKI_CONTENT = `
-> WarpEngine is a **standalone**, mountable Rails engine that turns any Rails application into a retro software catalog.
+// Generic markdown in the shape every engine-tagged wiki page must follow:
+// intro paragraph, then "What You Get" as the first heading with bullets.
+// Dummy text only — real content always comes from the wiki at runtime.
+const SAMPLE_MARKDOWN = `
+> Example Engine is a **sample** framework that turns markdown into landing cards.
 
 # What You Get
 
-- **Catalog domain**: \`Software\`, \`Release\` — all soft-deleted, with download tracking.
-- **CI pipeline server**: Woodpecker [configuration extension](https://woodpecker-ci.org/docs) — the engine serves the full pipeline.
+- **First feature**: some \`inline code\` — with a longer explanation.
+- **Second feature**: a [link](https://example.org) inside the text.
 - Plain bullet without a bold lead.
 
-# Endpoints
+# Later Section
 
 - **Not picked up**: a list under a later heading.
 `
 
 describe('getEngineDigest', () => {
   it('extracts the intro from the leading blockquote with inline markdown stripped', () => {
-    const digest = getEngineDigest(WIKI_CONTENT)
+    const digest = getEngineDigest(SAMPLE_MARKDOWN)
     expect(digest.intro).toBe(
-      'WarpEngine is a standalone, mountable Rails engine that turns any Rails application into a retro software catalog.',
+      'Example Engine is a sample framework that turns markdown into landing cards.',
     )
   })
 
-  it('takes the heading above the first bullet list as the highlights title', () => {
-    expect(getEngineDigest(WIKI_CONTENT).highlightsTitle).toBe('What You Get')
+  it('takes the What You Get heading as the highlights title', () => {
+    expect(getEngineDigest(SAMPLE_MARKDOWN).highlightsTitle).toBe('What You Get')
   })
 
   it('parses bold-lead bullets into title and text', () => {
-    const [first, second] = getEngineDigest(WIKI_CONTENT).highlights
+    const [first, second] = getEngineDigest(SAMPLE_MARKDOWN).highlights
     expect(first).toEqual({
-      title: 'Catalog domain',
-      text: 'Software, Release — all soft-deleted, with download tracking.',
+      title: 'First feature',
+      text: 'some inline code — with a longer explanation.',
+    })
+    expect(second).toEqual({
+      title: 'Second feature',
+      text: 'a link inside the text.',
     })
-    expect(second.title).toBe('CI pipeline server')
-    expect(second.text).toBe('Woodpecker configuration extension — the engine serves the full pipeline.')
   })
 
   it('keeps bullets without a bold lead as plain text', () => {
-    const third = getEngineDigest(WIKI_CONTENT).highlights[2]
+    const third = getEngineDigest(SAMPLE_MARKDOWN).highlights[2]
     expect(third).toEqual({ title: '', text: 'Plain bullet without a bold lead.' })
   })
 
-  it('stops at the heading after the first list', () => {
-    const digest = getEngineDigest(WIKI_CONTENT)
-    expect(digest.highlights).toHaveLength(3)
+  it('stops at the heading after the What You Get section', () => {
+    expect(getEngineDigest(SAMPLE_MARKDOWN).highlights).toHaveLength(3)
+  })
+
+  it('matches the heading case-insensitively', () => {
+    const digest = getEngineDigest('# WHAT you get\n- **A**: b.')
+    expect(digest.highlights).toEqual([{ title: 'A', text: 'b.' }])
+  })
+
+  it('yields no highlights when the first heading is not What You Get', () => {
+    const digest = getEngineDigest('Intro line.\n\n# Features\n- **A**: b.\n\n# What You Get\n- **C**: d.')
+    expect(digest.intro).toBe('Intro line.')
+    expect(digest.highlights).toEqual([])
   })
 
   it('caps the highlights at six items', () => {
-    const many = '# Features\n' + Array.from({ length: 9 }, (_, i) => `- item ${i}`).join('\n')
+    const many = '# What You Get\n' + Array.from({ length: 9 }, (_, i) => `- item ${i}`).join('\n')
     expect(getEngineDigest(many).highlights).toHaveLength(6)
   })
 
   it('uses the first paragraph as intro when there is no blockquote', () => {
-    const digest = getEngineDigest('Just a plain paragraph.\n\n---\n\n# Stuff\n- one')
+    const digest = getEngineDigest('Just a plain paragraph.\n\n---\n\n# What You Get\n- one')
     expect(digest.intro).toBe('Just a plain paragraph.')
     expect(digest.highlights).toEqual([{ title: '', text: 'one' }])
   })
diff --git a/apps/frontend/src/stores/engines.store.ts b/apps/frontend/src/stores/engines.store.ts
index 0515523..d6906f6 100644
--- a/apps/frontend/src/stores/engines.store.ts
+++ b/apps/frontend/src/stores/engines.store.ts
@@ -9,14 +9,16 @@ export interface EngineHighlight {
   text: string
 }
 
-// Landing-page digest of a wiki engine page: the intro paragraph plus the
-// first bullet list (e.g. "What You Get") parsed into feature highlights.
+// Landing-page digest of a wiki engine page. Convention: every engine-tagged
+// page opens with an intro paragraph, then a "What You Get" heading as its
+// first heading, with the feature bullets underneath.
 export interface EngineDigest {
   intro: string
   highlightsTitle: string
   highlights: EngineHighlight[]
 }
 
+const HIGHLIGHTS_HEADING = 'what you get'
 const MAX_HIGHLIGHTS = 6
 
 // Inline markdown (links, bold, code) stripped so the text reads as plain prose.
@@ -31,33 +33,38 @@ function stripInline(md: string): string {
 
 export function getEngineDigest(content: string): EngineDigest {
   const digest: EngineDigest = { intro: '', highlightsTitle: '', highlights: [] }
-  let heading = ''
+  let inHighlights = false
 
   for (const raw of (content || '').split('\n')) {
     const line = raw.trim()
     if (!line || /^([-*_])\1{2,}$/.test(line)) continue
 
-    if (/^#{1,6}\s/.test(line)) {
-      // A heading after the list means the highlight section is over.
-      if (digest.highlights.length) break
-      heading = stripInline(line.replace(/^#{1,6}\s*/, ''))
+    const heading = line.match(/^#{1,6}\s+(.*)$/)
+    if (heading) {
+      // The section ends at the next heading; and if the page's first heading
+      // is not "What You Get", it doesn't follow the convention — no highlights.
+      if (inHighlights) break
+      const title = stripInline(heading[1])
+      if (title.toLowerCase() !== HIGHLIGHTS_HEADING) break
+      inHighlights = true
+      digest.highlightsTitle = title
       continue
     }
 
-    if (/^[-*]\s/.test(line)) {
-      const item = line.replace(/^[-*]\s+/, '')
-      const lead = item.match(/^\*\*([^*]+)\*\*\s*[:—–-]?\s*(.*)$/)
+    if (inHighlights) {
+      const bullet = line.match(/^[-*]\s+(.*)$/)
+      if (!bullet) continue
+      const lead = bullet[1].match(/^\*\*([^*]+)\*\*\s*[:—–-]?\s*(.*)$/)
       digest.highlights.push(
         lead
           ? { title: stripInline(lead[1]), text: stripInline(lead[2]) }
-          : { title: '', text: stripInline(item) },
+          : { title: '', text: stripInline(bullet[1]) },
       )
-      if (!digest.highlightsTitle) digest.highlightsTitle = heading
       continue
     }
 
-    // First prose line (paragraph or blockquote) before any list is the intro.
-    if (!digest.intro && !digest.highlights.length) {
+    // First prose line (paragraph or blockquote) before the heading is the intro.
+    if (!digest.intro) {
       digest.intro = stripInline(line.replace(/^>\s*/, ''))
     }
   }