auto-git:

[add] scripts/.DS_Store
 [add] scripts/README.md
 [add] scripts/extract_texture_filename_from_3ds.py
 [add] scripts/generate_3d_glb.py
 [add] scripts/generate_json.py
 [add] scripts/image_from_json.py
 [add] scripts/naming.py
 [add] scripts/openai_image_gen.py
 [add] scripts/remesh_bake_batch.py
 [add] server/public/assets/images/.DS_Store
 [add] server/public/assets/models/spirits/.DS_Store
 [change] server/public/assets/.DS_Store
 [change] server/public/assets/models/.DS_Store

scripts/naming.py

@@ -0,0 +1,66 @@
+import os
+import json
+import re
+from difflib import get_close_matches
+
+# ---- Konfiguration ----
+image_dir = "webp"
+json_path = "spirit_list.json"
+output_path = "spirit_list_with_images.json"
+image_url_prefix = "/assets/images/spirits/"  # Deine URL
+
+# --- Hilfsfunktion: Normalisiere Namen (um sie vergleichbar zu machen) ---
+def norm(s):
+    s = s.lower()
+    s = re.sub(r'[^a-z0-9]+', '', s)  # Alles außer Buchstaben/Zahlen raus
+    return s
+
+# ---- Bilddateien einlesen & normalisieren ----
+image_files = [f for f in os.listdir(image_dir) if f.lower().endswith('.webp')]
+norm2file = {norm(os.path.splitext(f)[0]): f for f in image_files}
+
+# ---- JSON einlesen ----
+with open(json_path, "r", encoding="utf-8") as f:
+    spirits = json.load(f)
+
+matched = 0
+notfound = []
+
+for entry in spirits:
+    # Nimm zuerst Model URL, ansonsten Name
+    base = None
+    if "Model URL" in entry and entry["Model URL"]:
+        base = os.path.splitext(os.path.basename(entry["Model URL"]))[0]
+    if not base and "Name" in entry:
+        base = entry["Name"]
+    if not base:
+        notfound.append(entry)
+        continue
+
+    base_norm = norm(base)
+    # Direktes Mapping versuchen
+    if base_norm in norm2file:
+        entry["Image URL"] = image_url_prefix + norm2file[base_norm]
+        matched += 1
+        continue
+
+    # Fuzzy-Match, falls nicht gefunden
+    candidates = get_close_matches(base_norm, norm2file.keys(), n=1, cutoff=0.7)
+    if candidates:
+        file_name = norm2file[candidates[0]]
+        entry["Image URL"] = image_url_prefix + file_name
+        print(f"Fuzzy: {base} → {file_name}")
+        matched += 1
+    else:
+        print(f"Kein Bild gefunden für: {base}")
+        notfound.append(entry)
+
+# --- Neue JSON schreiben ---
+with open(output_path, "w", encoding="utf-8") as f:
+    json.dump(spirits, f, indent=2, ensure_ascii=False)
+
+print(f"{matched} von {len(spirits)} Einträgen mit Bild gematcht.")
+print(f"Nicht gefunden: {len(notfound)}")
+if notfound:
+    for entry in notfound:
+        print("  -", entry.get("Name", "???"))