diff --git a/scripts/OPNSense/pyfrc2g-ciso_assist.py b/scripts/OPNSense/pyfrc2g-ciso_assist.py
index e69de29..42b2ac6 100644
--- a/scripts/OPNSense/pyfrc2g-ciso_assist.py
+++ b/scripts/OPNSense/pyfrc2g-ciso_assist.py
@@ -0,0 +1,325 @@
+import requests
+import json
+import urllib3
+urllib3.disable_warnings(urllib3.exceptions.InsecureRequestWarning)
+import re
+from collections import OrderedDict
+from graphviz import Digraph
+import os
+import logging
+import glob
+import csv
+from config import INTERFACE_MAP, NET_MAP, ADDRESS_MAP, PORT_MAP
+import hashlib
+
+logging.basicConfig(level=logging.INFO)
+
+# --- CONFIG ---
+OPNS_URL = "https://<OPNS_ADDRESS/api/firewall/filter/search_rule"
+OPNS_SECRET = "<API_SECRET>"
+OPNS_KEY = "<API_KEY>"
+PASSERELLE = "GW_NAME"
+FICHIER_CSV = "output_"+PASSERELLE+".csv"
+GRAPH_OUTPUT_DIR = "tmp/graphs_"+PASSERELLE
+INTERFACES = ["wan","lan","opt1"]
+# CISO Assistant
+CISO_URL = "https://<CISO_ASSISTANT_ADDRESS>"
+CISO_TOKEN = "<CISO_ASSISTANT_TOKEN>"
+CISO_EVIDENCE = f"{CISO_URL}/api/evidences/<EVIDENCE_ID>/upload/"
+
+def md5sum(path):
+    md5 = hashlib.md5()
+    with open(path, "rb") as f:
+        # Lire le fichier par blocs pour éviter de saturer la mémoire
+        for chunk in iter(lambda: f.read(4096), b""):
+            md5.update(chunk)
+    return md5.hexdigest()
+
+def recup_regles(url, api_secret, api_key, params):
+    try:
+        #headers = {"accept": "application/json", "X-API-Key": token}
+        reponse = requests.get(
+            url,
+            params=params,
+            auth=(api_key, api_secret),   # même ordre que -u "SECRET:KEY"
+            verify=False                  # équivalent de curl -k
+        )
+        return reponse.json()
+    except ValueError:
+        print("Échec de la connexion:", reponse.status_code, reponse.text)
+        exit()
+
+def safe_value(value, field=None):
+    # if value is None:
+    #     print("c'est None")
+    #     return "Any"
+    if isinstance(value, list):
+        value = ", ".join(map(str, value))
+    if field in ("source", "interface"):
+        val = str(value).lower()
+        if val in INTERFACE_MAP:
+            return INTERFACE_MAP[val]
+    if str(field) in ("destination_port"):
+        val = str(value)
+        if val in PORT_MAP:
+            return PORT_MAP[val]
+    if str(field) in ("destination"):
+        val = str(value).lower()
+        if val in NET_MAP:
+            return NET_MAP[val]
+    if str(field) in ("destination"):
+        val = str(value).lower()
+        if val in ADDRESS_MAP:
+            return ADDRESS_MAP[val]
+    return value
+
+def normalize_ports(port_field):
+    if not port_field:
+        return "Any"
+    return re.sub(r'\s+', '', port_field.strip()) or "Any"
+
+def export_to_ciso(url,token,fichier):
+    upload_url = url
+    upload_headers = {
+        'Authorization': f'Token {token}',
+        'accept': 'application/json',
+        'Content-Type': 'document',
+        'Content-Disposition': f'attachment; filename={fichier}'
+    }
+    file_path = fichier
+    with open(file_path, 'rb') as file:
+        response = requests.post(upload_url, headers=upload_headers, data=file, verify=False)
+    if response.status_code == 200:
+        return True
+    else:
+        return False
+
+def parse_csv_and_generate(csv_path, output_dir):
+    os.makedirs(output_dir, exist_ok=True)
+    flux_par_passerelle = OrderedDict()
+    next_id = 0
+
+    def get_node(nodes_local, key, label=None, color=None, force_unique=False):
+        """Crée ou récupère un nœud factorisé par cluster/source sauf si force_unique."""
+        nonlocal next_id
+        actual_key = f"{key}__{next_id}" if force_unique else key
+        if actual_key not in nodes_local:
+            nodes_local[actual_key] = (f"node{next_id}", color, label if label else key)
+            next_id += 1
+        return nodes_local[actual_key][0]
+
+    def get_action_color(action):
+        return "#a3f7a3" if action == "PASS" else "#f7a3a3" if action == "BLOCK" else None
+
+    def get_destination_color(disabled):
+        return "#ffcc00" if disabled == "True" else None
+
+    with open(csv_path, newline='', encoding='utf-8') as f:
+        reader = csv.DictReader(f)
+        for row in reader:
+            source = (row.get("SOURCE") or "").strip()
+            passerelle = (row.get("PASSERELLE") or "").strip()
+            action = (row.get("ACTION") or "").strip().upper()
+            protocole = (row.get("PROTOCOLE") or "").strip() or "Any"
+            ports = normalize_ports(row.get("PORT"))
+            destination = (row.get("DESTINATION") or "").strip()
+            descr = (row.get("COMMENTAIRE") or "").strip()
+
+            source_label = f"SOURCE | {source}" if source else "SOURCE | <inconnu>"
+            passerelle_label = f"PASSERELLE | {passerelle}" if passerelle else "PASSERELLE | <inconnu>"
+            action_label = f"ACTION | {action}" if action else "ACTION | <inconnu>"
+            proto_label = f"PROTOCOLE | {protocole}"
+            port_label = f"PORT | {ports}"
+            destination_label = f"{destination} | {descr}" if descr else f" VLAN | {destination}" or "<inconnu>"
+
+            # --- Initialisation cluster/source ---
+            if passerelle not in flux_par_passerelle:
+                flux_par_passerelle[passerelle] = OrderedDict()
+            if source not in flux_par_passerelle[passerelle]:
+                flux_par_passerelle[passerelle][source] = {"nodes": OrderedDict(), "edges": set()}
+
+            cluster = flux_par_passerelle[passerelle][source]
+
+            # --- Création des nœuds ---
+            n_source = get_node(cluster["nodes"], source_label)
+            n_pass = get_node(cluster["nodes"], passerelle_label)
+            n_action = get_node(cluster["nodes"], action_label, color=get_action_color(action))
+            proto_key = f"{protocole}|{action}"
+            n_proto = get_node(cluster["nodes"], proto_key, label=proto_label)
+            port_key = f"{ports}|{proto_key}"
+            n_port = get_node(cluster["nodes"], port_key, label=port_label)
+            if "Regles-flottantes" in passerelle:
+                n_destination = get_node(cluster["nodes"], destination_label)
+            else:
+                n_destination = get_node(cluster["nodes"], destination_label, force_unique=True)  # DESTINATION non factorisée
+                
+            edges = [
+                (n_source, n_pass),
+                (n_pass, n_action),
+                (n_action, n_proto),
+                (n_proto, n_port),
+                (n_port, n_destination),
+            ]
+
+            cluster["edges"].update(edges)
+
+    # --- Génération des graphes ---
+    for passerelle, sources in flux_par_passerelle.items():
+        filename = os.path.join(output_dir, f"{passerelle.replace('/', '_')}.gv")
+        g = Digraph('g', filename=filename, format='png')
+        g.attr(fontname="Helvetica,Arial,sans-serif")
+        g.attr("node", fontname="Helvetica,Arial,sans-serif", fontsize="11", shape="record")
+        g.attr("edge", fontname="Helvetica,Arial,sans-serif")
+        g.attr(rankdir="LR")
+        g.attr(label=f"PASSERELLE : {passerelle}", labelloc="t", fontsize="14", color="#8888ff")
+
+        for source, cluster in sources.items():
+            with g.subgraph(name=f"cluster_{source.replace(' ', '_')}") as sg:
+                sg.attr(label=f"SOURCE : {source}", style="dashed", color="#aaaaaa")
+                for nid, color, label in cluster["nodes"].values():
+                    sg.node(nid, label=label, shape="record", **({"style":"filled","fillcolor":color} if color else {}))
+                for src, dst in cluster["edges"]:
+                    sg.edge(src, dst)
+
+        output_path = g.render(view=False)
+
+        # Suppression du fichier .gv après rendu
+        try:
+            if os.path.exists(filename):
+                os.remove(filename)
+                print(f"🗑️  Fichier temporaire supprimé : {filename}")
+        except Exception as e:
+            print(f"⚠️  Impossible de supprimer {filename} : {e}")
+
+        print(f"✅ Graph généré : {filename}.png")
+    
+    try:
+        from reportlab.pdfgen import canvas
+        from reportlab.lib.pagesizes import A4
+        from reportlab.lib.utils import ImageReader
+
+        # Récupération des PNG
+        png_files = sorted(glob.glob(os.path.join(output_dir, "*.png")))
+        if not png_files:
+            print("⚠️ Aucun fichier PNG trouvé pour le PDF.")
+            return
+
+        pdf_path = os.path.join(output_dir, PASSERELLE+"_MATRICE_DES_FLUX.pdf")
+
+        # Création PDF
+        c = canvas.Canvas(pdf_path, pagesize=A4)
+        width, height = A4
+
+        c.setTitle(f"Matrice des flux de la passerelle {PASSERELLE} ")
+
+        for i, png in enumerate(png_files):
+            # --- Titre / chapitre = nom du fichier ---
+            titre_page = os.path.basename(png).replace(".gv.png", "")
+
+            # Ajout du signet PDF
+            c.bookmarkPage(titre_page)
+            c.addOutlineEntry(titre_page, titre_page, level=0)
+
+            # Chargement de l'image
+            img = ImageReader(png)
+            img_width, img_height = img.getSize()
+
+            # Mise à l’échelle automatique
+            scale = min(width / img_width, height / img_height)
+            new_width = img_width * scale
+            new_height = img_height * scale
+
+            # Centrage
+            x = (width - new_width) / 2
+            y = (height - new_height) / 2
+
+            # Dessin
+            c.drawImage(img, x, y, width=new_width, height=new_height)
+
+            c.showPage()
+
+        c.save()
+        print(f"📄 PDF avec chapitres généré : {pdf_path}")
+
+        try:
+            for png in png_files:
+                if os.path.exists(png):
+                    os.remove(png)
+                    print(f"🗑️  PNG supprimé : {png}")
+        except Exception as e:
+            print(f"⚠️  Impossible de supprimer certains PNG : {e}")
+
+    except Exception as e:
+        print(f"⚠️ Erreur lors de la génération du PDF : {e}")
+
+    if not export_to_ciso(CISO_EVIDENCE, CISO_TOKEN, pdf_path):
+        logging.error("Échec de l'exportation dans Ciso.")
+        return
+
+# --- EXTRACTION DES DONNÉES ---
+with open(FICHIER_CSV, "w", newline="", encoding="utf-8") as f:
+    writer = csv.DictWriter(
+        f,
+        fieldnames=["SOURCE", "PASSERELLE", "ACTION", "PROTOCOLE", "PORT", "DESTINATION", "COMMENTAIRE"]
+    )
+    writer.writeheader()   # → On écrit l’entête une seule fois
+
+    # Boucle sur les interfaces
+    for interface in INTERFACES:
+        params = {
+            "interface": interface,
+            "show_all": "1"
+        }
+
+        data = recup_regles(OPNS_URL, OPNS_SECRET, OPNS_KEY, params)
+        entries = data.get("rows", [])
+
+        # Écriture des entrées
+        for entry in entries:
+            source_val = (
+                entry.get('source', {}).get('network')
+                or entry.get('source', {}).get('address')
+                or entry.get('source_net')
+                or entry.get('source', {}).get('any')
+            )
+            destination_val = (
+                entry.get('destination', {}).get('network')
+                or entry.get('destination', {}).get('address')
+                or entry.get('destination', {}).get('any')
+                or entry.get("destination_net")
+            )
+            port_dest_val = (
+                entry.get('destination', {}).get('port')
+                or entry.get("destination_port")
+            )
+            writer.writerow({
+                "SOURCE": safe_value(source_val, "source"),
+                "PASSERELLE": PASSERELLE + "/" + safe_value(entry.get("interface"), "interface")
+                                if entry.get("interface")
+                                else PASSERELLE + "/Regles-flottantes",
+                "ACTION": safe_value(entry.get("action")),
+                "PROTOCOLE": safe_value(entry.get("protocol")),
+                "PORT": safe_value(port_dest_val, "destination_port"),
+                "DESTINATION": safe_value(destination_val, "destination"),
+                "COMMENTAIRE": safe_value(entry.get("description"))
+            })
+
+
+# Récupération de la précédente somme md5sum
+with open("md5sum.txt", "r") as f:
+    prev_md5sum = f.readline().strip()  # .strip() enlève les retours à la ligne
+# Génération de la somme md5sum du fichier csv généré
+actual_md5sum = md5sum(FICHIER_CSV)
+
+# Comparaison des sommes md5sum. 
+# Si différentes => génération de la matrice.
+# Si identique => arrêt du script.
+if prev_md5sum != actual_md5sum:
+    with open("md5sum.txt", "w") as f:
+        f.write(actual_md5sum + "\n")
+    parse_csv_and_generate(FICHIER_CSV,GRAPH_OUTPUT_DIR)
+else:
+    logging.info("Pas de règles crées ou modifiées")
+
+if os.path.exists(FICHIER_CSV):
+    os.remove(FICHIER_CSV)
\ No newline at end of file