diff --git a/.gitea/workflows/test-skins.yml b/.gitea/workflows/test-skins.yml
index fd552ab9..1e2f53bb 100644
--- a/.gitea/workflows/test-skins.yml
+++ b/.gitea/workflows/test-skins.yml
@@ -14,7 +14,7 @@ jobs:
       - name: Validate links and assets
         shell: bash
         run: |
-          set -euo pipefail
+          set -uo pipefail
 
           RED="\033[31m"
           GREEN="\033[32m"
@@ -30,10 +30,12 @@ jobs:
               local url="$1"
               echo "      → Checking external: $url"
 
+              # HEAD request
               if curl -Is --max-time 10 "$url" | head -n1 | grep -qE "HTTP/.* (200|30[0-9])"; then
                   return 0
               fi
 
+              # GET fallback
               if curl -Is --max-time 10 -X GET "$url" | head -n1 | grep -qE "HTTP/.* (200|30[0-9])"; then
                   return 0
               fi
@@ -56,44 +58,38 @@ jobs:
           }
 
           extract_links() {
-              local file="$1"
+              local f="$1"
 
-              grep -oE '\[[^]]*\]\([^)]*\)' "$file" \
+              grep -oE '\[[^]]*\]\([^)]*\)' "$f" \
                   | sed -E 's/.*\((.*)\).*/\1/'
 
-              grep -oE '!\[[^]]*\]\([^)]*\)' "$file" \
+              grep -oE '!\[[^]]*\]\([^)]*\)' "$f" \
                   | sed -E 's/.*\((.*)\).*/\1/'
 
-              grep -oE 'https?://[^ )"]+' "$file"
+              grep -oE 'https?://[^ )"]+' "$f"
 
-              grep -oE '<img[^>]*src="[^"]+"' "$file" \
+              grep -oE '<img[^>]*src="[^"]+"' "$f" \
                   | sed -E 's/.*src="([^"]*)".*/\1/'
 
-              grep -oE '<video[^>]*src="[^"]+"' "$file" \
+              grep -oE '<video[^>]*src="[^"]+"' "$f" \
                   | sed -E 's/.*src="([^"]*)".*/\1/'
           }
 
           echo "🔍 Scanning Markdown files..."
           echo
 
-          while IFS= read -r mdfile; do
+          find . -type f -name '*.md' | while IFS= read -r mdfile; do
               echo "📄 Checking: $mdfile"
 
               while IFS= read -r url; do
                   [[ -z "$url" ]] && continue
                   [[ "$url" == mailto:* ]] && continue
 
-                  ###
-                  # 🚫 Skip tag-version links inside /docs/*
-                  ###
+                  # Skip tag links inside /docs → but **do NOT print them**
                   if [[ "$mdfile" == ./docs/* ]] && [[ "$url" == *"/src/tag/"* ]]; then
-                      echo "      → Skipping tag link (allowed in docs): $url"
                       continue
                   fi
 
-                  ###
-                  # Normal link checking
-                  ###
                   if [[ "$url" == http* ]]; then
                       if ! check_http "$url"; then
                           ERRORS+=("❌ Broken external link: $url (in $mdfile)")
@@ -107,18 +103,16 @@ jobs:
               done < <(extract_links "$mdfile")
 
               echo
-          done < <(find . -type f -name '*.md')
+          done
 
           echo
           if (( ${#ERRORS[@]} > 0 )); then
-              echo -e "${RED}✖ Broken Links & Missing Files Found:${RESET}"
-              for e in "${ERRORS[@]}"; do
-                  echo "$e"
-              done
+              echo -e "${RED}✖ Errors found:${RESET}"
+              printf "%s\n" "${ERRORS[@]}"
+
               echo
-              echo -e "${RED}⚠️ Job will NOT fail (requested behavior).${RESET}"
+              echo -e "${RED}❌ Failing job because broken links were found.${RESET}"
+              exit 1
           else
               echo -e "${GREEN}✔ All links OK!${RESET}"
           fi
-
-          exit 0