Add HTMLHint validation with auto-fix to all 10 news generation workflows

.github/workflows/news-article-generator.md

@@ -1123,6 +1123,9 @@ This validation checks:
 6. ✅ Index files have content (> 1KB) (blocking)
 7. ⚠️  Sitemap news-URL coverage (validated at build time; missing sitemap.xml is OK — it's generated by prebuild)
 8. ⚠️  Language switcher consistency across all 14 languages (warning level)
+9. ⚠️  Articles have article-top-nav with back-to-news link (warning level)
+10. ⚠️  Articles have back-to-news link in footer (warning level)
+11. ⚠️  HTMLHint validation on news articles (warning level; auto-fix attempted)
 
 **CRITICAL: Analysis Quality Check**
 Every generated article MUST contain real analysis, not merely a translated list of document links.
@@ -1165,13 +1168,20 @@ Before creating the PR, validate the quality of generated articles:
 
 **HTML Validation:**
 ```bash
-# Validate HTML structure
-npx htmlhint news/*.html --config .htmlhintrc
+# Validate HTML structure with auto-fix for common nesting errors
+if ! npx htmlhint "news/*-*.html" 2>/dev/null; then
+  echo "⚠️ HTML validation errors found, attempting auto-fix..."
+  npx tsx scripts/article-quality-enhancer.ts --fix
+  if ! npx htmlhint "news/*-*.html"; then
+    echo "❌ HTML validation errors remain after auto-fix. Please fix them before creating a PR."
+    exit 1
+  fi
+fi
 
 # Check for common issues:
 # - Missing alt attributes
 # - Duplicate IDs
-# - Invalid nesting
+# - Invalid nesting (<p><ul>, <p><div>)
 # - Missing required meta tags
 ```
 

.github/workflows/news-committee-reports.md

@@ -294,7 +294,30 @@ grep -o 'Why It Matters[^<]*' "news/$(date +%Y-%m-%d)-committee-reports-en.html"
 **Note**: News index files, metadata, and sitemap are generated automatically at build time by the `prebuild` script. Do NOT run generation scripts or commit their output — only commit the article HTML files. Run `npm run prebuild` (or `npm run build`) locally if you need to preview the generated indexes, metadata, or sitemap.
 
 ### Step 6: Validate & Create PR
-Validate HTML structure, then create PR:
+Run validation and HTMLHint before creating PR:
+```bash
+bash scripts/validate-news-generation.sh
+VALIDATION_EXIT=$?
+if [ "$VALIDATION_EXIT" -ne 0 ]; then
+  echo "❌ News generation validation failed. Fix the reported issues before creating a PR."
+  exit "$VALIDATION_EXIT"
+fi
+
+# HTMLHint validation with auto-fix
+NEWS_FILES=$(find news -maxdepth 1 -name '*-*.html' | wc -l)
+if [ "$NEWS_FILES" -gt 0 ]; then
+  if ! npx htmlhint "news/*-*.html" 2>/dev/null; then
+    echo "⚠️ HTML validation errors found, attempting auto-fix..."
+    npx tsx scripts/article-quality-enhancer.ts --fix
+    if ! npx htmlhint "news/*-*.html"; then
+      echo "❌ HTML validation errors remain after auto-fix. Please fix them before creating a PR."
+      exit 1
+    fi
+  fi
+fi
+```
+
+Then create PR:
 ```
 safeoutputs___create_pull_request
 ```

.github/workflows/news-evening-analysis.md

@@ -1001,9 +1001,28 @@ This validation checks:
 6. ✅ Index files have content (> 1KB) (blocking)
 7. ⚠️  Sitemap news-URL coverage (validated at build time; missing sitemap.xml is OK — it's generated by prebuild)
 8. ⚠️  Language switcher consistency across all 14 languages (warning level)
+9. ⚠️  Articles have article-top-nav with back-to-news link (warning level)
+10. ⚠️  Articles have back-to-news link in footer (warning level)
+11. ⚠️  HTMLHint validation on news articles (warning level; auto-fix attempted)
 
 **Exit code 0** = pass (proceed to Step 7), **exit code 1** = fail (STOP, do not create PR).
 
+**HTMLHint Auto-Fix:**
+After running `validate-news-generation.sh`, run HTMLHint with auto-fix for any remaining nesting errors:
+```bash
+NEWS_FILES=$(find news -maxdepth 1 -name '*-*.html' | wc -l)
+if [ "$NEWS_FILES" -gt 0 ]; then
+  if ! npx htmlhint "news/*-*.html" 2>/dev/null; then
+    echo "⚠️ HTML validation errors found, attempting auto-fix..."
+    npx tsx scripts/article-quality-enhancer.ts --fix
+    if ! npx htmlhint "news/*-*.html"; then
+      echo "❌ HTML validation errors remain after auto-fix. Please fix them before creating a PR."
+      exit 1
+    fi
+  fi
+fi
+```
+
 **CRITICAL: Analysis Quality Check**
 Every generated article MUST contain real analysis, not merely a translated list of document links.
 Verify each article includes analytical lede paragraphs, thematic grouping with commentary, policy significance analysis, and forward-looking "What to Watch" sections.

.github/workflows/news-month-ahead.md

@@ -232,6 +232,29 @@ npx tsx scripts/fix-article-navigation.ts
 
 ### Step 4: Translate, Validate & Verify Analysis Quality
 
+Run validation and HTMLHint before creating PR:
+```bash
+bash scripts/validate-news-generation.sh
+VALIDATION_EXIT=$?
+if [ "$VALIDATION_EXIT" -ne 0 ]; then
+  echo "❌ News generation validation failed. Fix the reported issues before creating a PR."
+  exit "$VALIDATION_EXIT"
+fi
+
+# HTMLHint validation with auto-fix for common nesting errors
+NEWS_FILES=$(find news -maxdepth 1 -name '*-*.html' | wc -l)
+if [ "$NEWS_FILES" -gt 0 ]; then
+  if ! npx htmlhint "news/*-*.html" 2>/dev/null; then
+    echo "⚠️ HTML validation errors found, attempting auto-fix..."
+    npx tsx scripts/article-quality-enhancer.ts --fix
+    if ! npx htmlhint "news/*-*.html"; then
+      echo "❌ HTML validation failed after auto-fix. Please fix remaining issues before creating PR."
+      exit 1
+    fi
+  fi
+fi
+```
+
 **CRITICAL: Each article MUST contain real analysis, not just a list of translated event titles.**
 Every generated article must include strategic outlook with political context, not merely translated calendar entries.
 

.github/workflows/news-monthly-review.md

@@ -240,6 +240,29 @@ npx tsx scripts/fix-article-navigation.ts
 
 ### Step 4: Translate, Validate & Verify Analysis Quality
 
+Run validation and HTMLHint before creating PR:
+```bash
+bash scripts/validate-news-generation.sh
+VALIDATION_EXIT=$?
+if [ "$VALIDATION_EXIT" -ne 0 ]; then
+  echo "❌ News generation validation failed. Fix the reported issues before creating a PR."
+  exit "$VALIDATION_EXIT"
+fi
+
+# HTMLHint validation with auto-fix for common nesting errors
+NEWS_FILES=$(find news -maxdepth 1 -name '*-*.html' | wc -l)
+if [ "$NEWS_FILES" -gt 0 ]; then
+  if ! npx htmlhint "news/*-*.html" 2>/dev/null; then
+    echo "⚠️ HTML validation errors found, attempting auto-fix..."
+    npx tsx scripts/article-quality-enhancer.ts --fix
+    if ! npx htmlhint "news/*-*.html"; then
+      echo "❌ HTML validation still failing after auto-fix. Please fix remaining issues manually before creating PR."
+      exit 1
+    fi
+  fi
+fi
+```
+
 **CRITICAL: Each article MUST contain real analysis, not just a list of translated document links.**
 Every generated article must include thematic analysis grouping documents by type and policy area, interpretive commentary on what the month's activity reveals about political dynamics, and key takeaways.
 

.github/workflows/news-motions.md

@@ -231,6 +231,29 @@ npx tsx scripts/fix-article-navigation.ts
 
 ### Step 4: Translate, Validate & Verify Analysis Quality
 
+Run validation and HTMLHint before creating PR:
+```bash
+bash scripts/validate-news-generation.sh
+VALIDATION_EXIT=$?
+if [ "$VALIDATION_EXIT" -ne 0 ]; then
+  echo "❌ News generation validation failed. Fix the reported issues before creating a PR."
+  exit "$VALIDATION_EXIT"
+fi
+
+# HTMLHint validation with auto-fix for common nesting errors
+NEWS_FILES=$(find news -maxdepth 1 -name '*-*.html' | wc -l)
+if [ "$NEWS_FILES" -gt 0 ]; then
+  if ! npx htmlhint "news/*-*.html" 2>/dev/null; then
+    echo "⚠️ HTML validation errors found, attempting auto-fix..."
+    npx tsx scripts/article-quality-enhancer.ts --fix
+    if ! npx htmlhint "news/*-*.html"; then
+      echo "❌ HTML validation errors remain after auto-fix. Please fix them before creating a PR."
+      exit 1
+    fi
+  fi
+fi
+```
+
 **CRITICAL: Each article MUST contain real analysis, not just a list of translated links.**
 Every generated article must include:
 - An analytical lede paragraph about opposition strategy and political fault lines (not just a motion count)

.github/workflows/news-propositions.md

@@ -230,6 +230,29 @@ npx tsx scripts/fix-article-navigation.ts
 
 ### Step 4: Translate, Validate & Verify Analysis Quality
 
+Run validation and HTMLHint before creating PR:
+```bash
+bash scripts/validate-news-generation.sh
+VALIDATION_EXIT=$?
+if [ "$VALIDATION_EXIT" -ne 0 ]; then
+  echo "❌ News generation validation failed. Fix the reported issues before creating a PR."
+  exit "$VALIDATION_EXIT"
+fi
+
+# HTMLHint validation with auto-fix for common nesting errors
+NEWS_FILES=$(find news -maxdepth 1 -name '*-*.html' | wc -l)
+if [ "$NEWS_FILES" -gt 0 ]; then
+  if ! npx htmlhint "news/*-*.html" 2>/dev/null; then
+    echo "⚠️ HTML validation errors found, attempting auto-fix..."
+    npx tsx scripts/article-quality-enhancer.ts --fix
+    if ! npx htmlhint "news/*-*.html"; then
+      echo "❌ HTML validation failed after auto-fix. Please resolve remaining HTMLHint errors before creating a PR."
+      exit 1
+    fi
+  fi
+fi
+```
+
 Translate all Swedish content, regenerate indexes, validate, then create PR.
 
 **CRITICAL: Each article MUST contain real analysis, not just a list of translated links.**

.github/workflows/news-realtime-monitor.md

@@ -800,9 +800,28 @@ This validation checks:
 6. ℹ️  Index files have content (skipped — index files generated at build time)
 7. ⚠️  Sitemap news-URL coverage (warning; missing sitemap.xml is OK — generated by prebuild)
 8. ⚠️  Language switcher consistency across all 14 languages (warning level)
+9. ⚠️  Articles have article-top-nav with back-to-news link (warning level)
+10. ⚠️  Articles have back-to-news link in footer (warning level)
+11. ⚠️  HTMLHint validation on news articles (warning level; auto-fix attempted)
 
 **Exit code 0** = all checks pass. **Exit code 1** = errors found. Both are recoverable — check elapsed time before deciding.
 
+**HTMLHint Auto-Fix:**
+After running `validate-news-generation.sh`, run HTMLHint with auto-fix for any remaining nesting errors:
+```bash
+NEWS_FILES=$(find news -maxdepth 1 -name '*-*.html' | wc -l)
+if [ "$NEWS_FILES" -gt 0 ]; then
+  if ! npx htmlhint "news/*-*.html" 2>/dev/null; then
+    echo "⚠️ HTML validation errors found, attempting auto-fix..."
+    npx tsx scripts/article-quality-enhancer.ts --fix
+    if ! npx htmlhint "news/*-*.html"; then
+      echo "❌ HTML validation errors remain after auto-fix. Please fix them before creating a PR."
+      exit 1
+    fi
+  fi
+fi
+```
+
 If validation shows errors, try to fix them. If elapsed >= 38 minutes, proceed to create PR with available articles.
 
 ### Step 6: Create PR (if articles generated)

.github/workflows/news-week-ahead.md

@@ -230,6 +230,29 @@ npx tsx scripts/fix-article-navigation.ts
 
 ### Step 4: Translate, Validate & Verify Analysis Quality
 
+Run validation and HTMLHint before creating PR:
+```bash
+bash scripts/validate-news-generation.sh
+VALIDATION_EXIT=$?
+if [ "$VALIDATION_EXIT" -ne 0 ]; then
+  echo "❌ News generation validation failed. Fix the reported issues before creating a PR."
+  exit "$VALIDATION_EXIT"
+fi
+
+# HTMLHint validation with auto-fix for common nesting errors
+NEWS_FILES=$(find news -maxdepth 1 -name '*-*.html' | wc -l)
+if [ "$NEWS_FILES" -gt 0 ]; then
+  if ! npx htmlhint "news/*-*.html" 2>/dev/null; then
+    echo "⚠️ HTML validation errors found, attempting auto-fix..."
+    npx tsx scripts/article-quality-enhancer.ts --fix
+    if ! npx htmlhint "news/*-*.html"; then
+      echo "❌ HTML validation failed after auto-fix. Please fix remaining HTMLHint errors before creating a PR."
+      exit 1
+    fi
+  fi
+fi
+```
+
 **CRITICAL: Each article MUST contain real analysis, not just a list of translated event titles.**
 Every generated article must include:
 - A "Why This Week Matters" context box with political significance analysis

.github/workflows/news-weekly-review.md

@@ -235,6 +235,29 @@ npx tsx scripts/fix-article-navigation.ts
 
 ### Step 4: Translate, Validate & Verify Analysis Quality
 
+Run validation and HTMLHint before creating PR:
+```bash
+bash scripts/validate-news-generation.sh
+VALIDATION_EXIT=$?
+if [ "$VALIDATION_EXIT" -ne 0 ]; then
+  echo "❌ News generation validation failed. Fix the reported issues before creating a PR."
+  exit "$VALIDATION_EXIT"
+fi
+
+# HTMLHint validation with auto-fix for common nesting errors
+NEWS_FILES=$(find news -maxdepth 1 -name '*-*.html' | wc -l)
+if [ "$NEWS_FILES" -gt 0 ]; then
+  if ! npx htmlhint "news/*-*.html" 2>/dev/null; then
+    echo "⚠️ HTML validation errors found, attempting auto-fix..."
+    npx tsx scripts/article-quality-enhancer.ts --fix
+    if ! npx htmlhint "news/*-*.html"; then
+      echo "❌ HTML validation failed after auto-fix. Please fix remaining issues before creating PR."
+      exit 1
+    fi
+  fi
+fi
+```
+
 **CRITICAL: Each article MUST contain real analysis, not just a list of translated document links.**
 Every generated article must include thematic analysis grouping documents by type and policy area, interpretive commentary on what the week's activity reveals about political dynamics, and key takeaways.
 

scripts/article-quality-enhancer.ts

@@ -423,6 +423,81 @@ export async function batchEnhanceQuality(
   return results;
 }
 
+/**
+ * Fix common HTML nesting errors in article content.
+ *
+ * Corrects invalid block-level nesting patterns:
+ * - `<p><ul>` → removes the enclosing `<p>` so the `<ul>` list is not nested inside a `<p>` paragraph
+ * - `<p><ol>` → removes the enclosing `<p>` so the `<ol>` list is not nested inside a `<p>` paragraph
+ * - `<p><div>` → removes the enclosing `<p>` so the block-level `<div>` is not nested inside a `<p>` paragraph
+ * - Orphaned `</p>` immediately after `</ul>` → removed
+ * - Orphaned `</p>` immediately after `</ol>` → removed
+ * - Orphaned `</p>` immediately after `</div>` (when wrapped by `<p>`) → removed
+ *
+ * @param content - Raw HTML content
+ * @returns Fixed HTML content
+ */
+export function fixArticleHtmlNesting(content: string): string {
+  let fixed = content;
+
+  // Fix <p><ul>: remove the enclosing <p> so the list is not nested inside a paragraph
+  fixed = fixed.replace(/<p([^>]*)>\s*(<ul[\s>])/g, '$2');
+
+  // Fix <p><ol>: remove the enclosing <p> so the ordered list is not nested inside a paragraph
+  fixed = fixed.replace(/<p([^>]*)>\s*(<ol[\s>])/g, '$2');
+
+  // Fix <p><div>: remove enclosing <p> around block-level div
+  fixed = fixed.replace(/<p([^>]*)>\s*(<div[\s>])/g, '$2');
+
+  // Fix orphaned </p> immediately after </ul>
+  fixed = fixed.replace(/<\/ul>\s*<\/p>/g, '</ul>');
+
+  // Fix orphaned </p> immediately after </ol>
+  fixed = fixed.replace(/<\/ol>\s*<\/p>/g, '</ol>');
+
+  // Fix orphaned </p> immediately after </div> that was wrapped in <p>
+  fixed = fixed.replace(/<\/div>\s*<\/p>/g, '</div>');
+
+  return fixed;
+}
+
+/**
+ * Fix HTML nesting errors in a file in-place.
+ *
+ * @param filePath - Path to the HTML file to fix
+ * @returns True if the file was modified
+ */
+export function fixHtmlNestingInFile(filePath: string): boolean {
+  if (!fs.existsSync(filePath)) {
+    return false;
+  }
+  const original = fs.readFileSync(filePath, 'utf-8');
+  const fixed = fixArticleHtmlNesting(original);
+  if (fixed !== original) {
+    fs.writeFileSync(filePath, fixed, 'utf-8');
+    return true;
+  }
+  return false;
+}
+
+// CLI entry point: support --fix flag to fix HTML nesting in news/*-*.html article files
+if (import.meta.url === `file://${process.argv[1]}`) {
+  if (process.argv.includes('--fix')) {
+    const globPattern = process.argv[process.argv.indexOf('--fix') + 1];
+    const targetGlob = (globPattern && !globPattern.startsWith('-')) ? globPattern : 'news/*-*.html';
+    const { globSync } = await import('glob');
+    const files: string[] = globSync(targetGlob);
+    let fixedCount = 0;
+    for (const file of files) {
+      if (fixHtmlNestingInFile(file)) {
+        console.log(`Fixed HTML nesting in: ${file}`);
+        fixedCount++;
+      }
+    }
+    console.log(`Fixed ${fixedCount} of ${files.length} files.`);
+  }
+}
+
 // Export individual assessment functions for testing
 export {
   assessAnalyticalDepth,