add some things

Author: SyntaxError4Life

README.md

@@ -48,6 +48,19 @@ gr.ChatInterface(
 ).launch()
 ```
 
+---
+
+## Use the commands :
+
+- `taranis help` : in the name...
+- `taranis update` : upgrade the framework
+- `taranis open` : open the TUI
+
+### The TUI :
+![TUI](images/TUI.png)
+
+- `/help` to start
+
 ## Documentation :
 
 - [Base of the docs](https://zanomega.com/open-taranis/) (coding some things before the real docs)
@@ -58,11 +71,7 @@ gr.ChatInterface(
 - [X]   v0.0.x: Add and confirm other API providers (in the cloud, not locally)
 - [X]   v0.1.x: Functionality verifications in [examples](https://github.com/SyntaxError4Life/open-taranis/blob/main/examples/)
 - [ ] > v0.2.0: Add features for **logic-only coding** approach
-- [ ]   v0.6.x: Add llama.cpp as backend in addition to APIs
-- [ ]   v0.7.x: Add reverse proxy + server to create a dedicated full relay/backend (like OpenRouter), framework usable as server and client
-- [ ]   v0.8.x: Add PyTorch as backend with `transformers` to deploy a remote server
-- [ ]   v0.9.x: Total reduction of dependencies for built-in functions (unless counter-optimizations)
-- [ ]   v1.0.0: First complete version in Python without dependencies
+- The rest will follow soon.
 
 ## Changelog
 
@@ -73,6 +82,7 @@ gr.ChatInterface(
 - **v0.1.2** : Fixed a display bug in the **web_front** and experimentally added **ollama as a backend**
 - **v0.1.3** : Fixed the memory reset in the **web_front** and remove **ollama module** for **openai front** (work 100 times better)
 - **v0.1.4** : Fixed `web_front` for native use on huggingface, as well as `handle_streaming` which had tool retrieval issues
+- **v0.1.5** : Added a **TUI** and **commands**, detection of **env variables** (API keys) and tools in the framework
 
 ## Advanced Examples
 

src/open_taranis/CLI.py

@@ -18,29 +18,26 @@
 ████  █████  ████  ██  ███  ███  ████  ██  ██    █████  ███████████  █
 ████  █████  ████  ██  ████  ██  ████  ██  ███   ██        ███      ██"""
 
-# Contraintes minimales
+
 MIN_HEIGHT = 24
 MIN_WIDTH = 80
 
 def run(stdscr):
-    # Configuration basique de curses
-    curses.curs_set(1) # Curseur visible
+    curses.curs_set(1)
     curses.start_color()
     curses.use_default_colors()
 
-    # Initialisation des paires de couleurs
-    curses.init_pair(1, curses.COLOR_RED, -1)   # Pour le logo
-    curses.init_pair(2, curses.COLOR_WHITE, -1) # Pour le texte standard
-    curses.init_pair(3, curses.COLOR_YELLOW, -1)# Pour les erreurs
+    curses.init_pair(1, curses.COLOR_RED, -1)
+    curses.init_pair(2, curses.COLOR_WHITE, -1)
+    curses.init_pair(3, curses.COLOR_YELLOW, -1)
 
     input_buffer = ""
-    display_mode = "NONE" # États possibles : "NONE", "HELP", "API", etc.
+    display_mode = "NONE"
+    text = []
 
     while True:
-        # Récupération dynamique des dimensions du terminal
         height, width = stdscr.getmaxyx()
 
-        # Vérification des contraintes minimales
         if height < MIN_HEIGHT or width < MIN_WIDTH:
             stdscr.clear()
             msg = f"Terminal too small: {width}x{height} (min {MIN_WIDTH}x{MIN_HEIGHT})"
@@ -55,12 +52,7 @@ def run(stdscr):
                 break
             continue
 
-        # Nettoyage de l'écran pour le rafraîchissement
         stdscr.clear()
-        
-        # --- Calcul dynamique du Layout ---
-        
-        # 1. Traitement du Logo (Haut Gauche)
         logo_lines = LOGO_ASCII.split('\n')
         logo_height = len(logo_lines)
 
@@ -72,11 +64,9 @@ def run(stdscr):
                 except curses.error:
                     pass
 
-        # 2. Zone de Contenu (Centre)
         content_start = logo_height + 1
         content_end = height - 3
 
-        # --- Affichage du contenu central selon le mode ---
         if display_mode == "HELP":
             text = [
                 "Commands :",
@@ -100,7 +90,7 @@ def run(stdscr):
 
         elif display_mode == 'MORE_API':
             text = [
-                "APIs registered :",
+                "APIs and env_var",
                 "- openrouter  = 'OPENROUTER_API'",
                 "- huggingface = 'HF_API'",
                 "- venice.ai   = 'VENICEAI_API'",
@@ -119,8 +109,6 @@ def run(stdscr):
                         pass
                     current_line += 1
 
-        
-        # 3. Footer / Invite de commande (Bas)
         sep_y = height - 2
         input_y = height - 1
 
@@ -141,13 +129,12 @@ def run(stdscr):
 
         stdscr.refresh()
 
-        # --- Gestion des entrées clavier ---
         key = stdscr.getch()
 
         if key == curses.KEY_RESIZE:
             continue
 
-        elif key in (10, 13): # Entrée
+        elif key in (10, 13):
             command = input_buffer.strip()
 
             if command == "/exit":
@@ -163,18 +150,17 @@ def run(stdscr):
                 display_mode = 'MORE_API'
 
             else:
-                # Commande invalide ou vide : on efface l'affichage central
                 display_mode = None
 
             input_buffer = ""
 
-        elif key in (127, curses.KEY_BACKSPACE, ord('\b')): # Backspace
+        elif key in (127, curses.KEY_BACKSPACE, ord('\b')):
             input_buffer = input_buffer[:-1]
 
-        elif key == 27: # Échap
+        elif key == 27:
             input_buffer = ""
 
-        elif 32 <= key <= 126: # Caractères imprimables ASCII
+        elif 32 <= key <= 126:
             input_buffer += chr(key)
 
 # ==============================
@@ -193,13 +179,11 @@ def main():
 """)
 
     elif argv[1] == "open":
-        # Lancement de la boucle curses
         curses.wrapper(run)
 
     elif argv[1] == "update":
         print("Updating open-taranis via pip...")
         try:
-            # On lance pip install -U sur le paquet actuel
             subprocess.check_call([sys.executable, "-m", "pip", "install", "-U", "open-taranis"])
             print("Update successful.")
         except subprocess.CalledProcessError as e: