RC version

- AI & Session Management:
  - Add random suffix to session IDs to ensure uniqueness.
  - Implement optional pagination/limit in session listing (default 20).
  - Add `--all` flag to `ai` CLI commands to view all sessions.
  - Keep active session ID and path synced correctly during clean session startups.

- CLI UI/UX:
  - Add a custom `ConnpyTheme` for inquirer prompts that dynamically translates
    active hex style colors to terminal ANSI/blessed escapes.

- gRPC & Services:
  - Implement remote MCP server listing (`list_mcp_servers` RPC and services).
  - Stream responder updates (`__RESPONDER__`) to toggle headers dynamically between
    "Network Engineer" and "Network Architect" on remote/web clients.
  - Fix remote client deadlock risk by ensuring `final_mark` is sent on exceptions.
  - Hydrate client-side chat history correctly on initial streaming request.

- Testing:
  - Add integration tests for AI gRPC services and MCP server listing.

.gitignore

@@ -146,11 +146,13 @@ package.json
 
 # Development docs
 connpy_roadmap.md
+testnew/
 testall/
 testremote/
 *.db
 *.patch
 scratch.py
+connpy.code-workspace
 
 # Internal planning and implementation docs
 PLAN_CAPA_SERVICIOS.md
@@ -170,3 +172,6 @@ MULTI_USER_IMPLEMENTATION_STEPS.md
 #themes
 nord.yml
 theme.py
+
+#ai auth
+auth.json

README.md

@@ -17,8 +17,12 @@ The v6 release introduces the **AI Copilot**, an interactive terminal assistant
 ## 🤖 AI Copilot (New in v6)
 The AI Copilot is deeply integrated into your terminal workflow:
 - **Terminal Context Awareness**: The Copilot can "see" your screen output, helping you diagnose errors or analyze command results in real-time.
+- **Dynamic Context Selection**: Flexibly select single, range, or line-based terminal blocks to feed the Copilot, filtering out interactive scrolling garbage automatically (e.g., Cisco IOS/XR scrolling, paginators).
 - **Hybrid Multi-Agent System**: Automatically escalates complex tasks between the **Network Engineer** (execution) and the **Network Architect** (strategy).
 - **MCP Integration**: Dynamically load tools from external providers (6WIND, AWS, etc.) via the Model Context Protocol.
+- **Flexible Auth & Keyless AI**: Support for advanced LiteLLM credentials (`--engineer-auth` / `--architect-auth`) allowing keyless local models (Ollama), cloud engines (Vertex AI), or custom endpoints.
+- **Enhanced Session Management**: Uniquely generated sessions, robust pagination, and interactive styling translating prompt themes directly to terminal escapes.
+- **Semantic Prompt Integration**: Emit standard OSC prompt sequences (`\x1b]133;B`) for real-time remote/web front-end command tracking.
 - **Interactive Chat**: Launch with `conn ai` for a collaborative troubleshooting session.
 
 

connpy/__init__.py

@@ -19,8 +19,12 @@ The v6 release introduces the **AI Copilot**, an interactive terminal assistant
 ## 🤖 AI Copilot (New in v6)
 The AI Copilot is deeply integrated into your terminal workflow:
 - **Terminal Context Awareness**: The Copilot can "see" your screen output, helping you diagnose errors or analyze command results in real-time.
+- **Dynamic Context Selection**: Flexibly select single, range, or line-based terminal blocks to feed the Copilot, filtering out interactive scrolling garbage automatically (e.g., Cisco IOS/XR scrolling, paginators).
 - **Hybrid Multi-Agent System**: Automatically escalates complex tasks between the **Network Engineer** (execution) and the **Network Architect** (strategy).
 - **MCP Integration**: Dynamically load tools from external providers (6WIND, AWS, etc.) via the Model Context Protocol.
+- **Flexible Auth & Keyless AI**: Support for advanced LiteLLM credentials (`--engineer-auth` / `--architect-auth`) allowing keyless local models (Ollama), cloud engines (Vertex AI), or custom endpoints.
+- **Enhanced Session Management**: Uniquely generated sessions, robust pagination, and interactive styling translating prompt themes directly to terminal escapes.
+- **Semantic Prompt Integration**: Emit standard OSC prompt sequences (`\x1b]133;B`) for real-time remote/web front-end command tracking.
 - **Interactive Chat**: Launch with `conn ai` for a collaborative troubleshooting session.
 
 
@@ -203,5 +207,7 @@ __pdoc__ = {
     'nodes.deferred_class_hooks': False,
     'connapp': False,
     'connapp.encrypt': True,
-    'printer': False
+    'printer': False,
+    'tests': False
 }
+

connpy/_version.py

@@ -1 +1 @@
-__version__ = "6.0.0b11"
+__version__ = "6.0.0b12"

connpy/ai.py

@@ -1,4 +1,6 @@
 import os
+import secrets
+
 import sys
 import json
 import re
@@ -106,7 +108,7 @@ class ai:
         r'^systemctl\s+status\s+', r'^journalctl\s+'
     ]
 
-    def __init__(self, config, org=None, api_key=None, engineer_model=None, architect_model=None, engineer_api_key=None, architect_api_key=None, console=None, confirm_handler=None, trust=False):
+    def __init__(self, config, org=None, api_key=None, engineer_model=None, architect_model=None, engineer_api_key=None, architect_api_key=None, console=None, confirm_handler=None, trust=False, engineer_auth=None, architect_auth=None, **kwargs):
         self.config = config
         self.console = console or printer.console
         self.confirm_handler = confirm_handler or self._local_confirm_handler
@@ -125,6 +127,29 @@ class ai:
         self.engineer_key = engineer_api_key or aiconfig.get("engineer_api_key")
         self.architect_key = architect_api_key or aiconfig.get("architect_api_key")
 
+        # Auth configurations (Prioridad: Argumento -> Config)
+        self.engineer_auth = engineer_auth if engineer_auth is not None else aiconfig.get("engineer_auth")
+        if self.engineer_auth is None:
+            self.engineer_auth = {}
+        elif not isinstance(self.engineer_auth, dict):
+            self.engineer_auth = {}
+
+        self.architect_auth = architect_auth if architect_auth is not None else aiconfig.get("architect_auth")
+        if self.architect_auth is None:
+            self.architect_auth = {}
+        elif not isinstance(self.architect_auth, dict):
+            self.architect_auth = {}
+
+        # Backward compatibility fallbacks: only inject api_key if the auth dict is empty/not configured
+        if self.engineer_key and not self.engineer_auth:
+            self.engineer_auth["api_key"] = self.engineer_key
+        if self.architect_key and not self.architect_auth:
+            self.architect_auth["api_key"] = self.architect_key
+
+        # Strategic Reasoning Engine (Architect) availability
+        is_architect_keyless = "vertex" in self.architect_model.lower() or "ollama" in self.architect_model.lower() or "local" in self.architect_model.lower()
+        self.has_architect = bool(self.architect_key or self.architect_auth or is_architect_keyless)
+
         # Custom Trusted Commands Regexes
         custom_trusted = aiconfig.get("trusted_commands", [])
         if isinstance(custom_trusted, str):
@@ -165,12 +190,12 @@ class ai:
         # Session Management
         self.sessions_dir = os.path.join(self.config.defaultdir, "ai_sessions")
         os.makedirs(self.sessions_dir, exist_ok=True)
-        self.session_id = None
+        self.session_id = getattr(self.config, "session_id", None)
-        self.session_path = None
+        self.session_path = os.path.join(self.sessions_dir, f"{self.session_id}.json") if self.session_id else None
 
         # Prompts base agnósticos
         architect_instructions = ""
-        if self.architect_key:
+        if self.has_architect:
             architect_instructions = """
             CRITICAL - CONSULT vs ESCALATE:
             - ALWAYS use 'consult_architect' for: Configuration planning, design decisions, complex troubleshooting.
@@ -186,7 +211,7 @@ class ai:
         else:
             architect_instructions = """
             CRITICAL - ARCHITECT UNAVAILABLE:
-            - The Strategic Reasoning Engine (Architect) is currently UNAVAILABLE because its API key is not configured.
+            - The Strategic Reasoning Engine (Architect) is currently UNAVAILABLE because its API key or authentication is not configured.
             - DO NOT attempt to consult or escalate to the architect.
             - If the user asks to consult the architect, inform them that the Architect is offline and offer to help them directly to the best of your abilities.
 """
@@ -292,15 +317,19 @@ class ai:
         if status_formatter:
             self.tool_status_formatters[name] = status_formatter
 
-    def _stream_completion(self, model, messages, tools, api_key, status=None, label="", debug=False, chunk_callback=None, **kwargs):
+    def _stream_completion(self, model, messages, tools, api_key=None, status=None, label="", debug=False, chunk_callback=None, auth=None, **kwargs):
         """Stream a completion call, rendering styled Markdown in real-time.
 
         Returns (response, streamed) where:
         - response: reconstructed ModelResponse (same as non-streaming)
         - streamed: True if text was rendered to console during streaming
         """
+        auth_dict = auth if auth is not None else {}
+        if api_key and "api_key" not in auth_dict:
+            auth_dict = auth_dict.copy()
+            auth_dict["api_key"] = api_key
 
-        stream_resp = completion(model=model, messages=messages, tools=tools, api_key=api_key, stream=True, **kwargs)
+        stream_resp = completion(model=model, messages=messages, tools=tools, stream=True, **auth_dict, **kwargs)
 
         chunks = []
         full_content = ""
@@ -743,7 +772,7 @@ class ai:
                 
                 try:
                     safe_messages = self._sanitize_messages(messages)
-                    response = completion(model=self.engineer_model, messages=safe_messages, tools=tools, api_key=self.engineer_key)
+                    response = completion(model=self.engineer_model, messages=safe_messages, tools=tools, **self.engineer_auth)
                 except Exception as e:
                     if status: status.stop()
                     raise ValueError(f"Engineer failed to connect: {str(e)}")
@@ -877,16 +906,27 @@ class ai:
                     continue
         return sorted(sessions, key=lambda x: x["created_at"], reverse=True)
 
-    def list_sessions(self):
+    def list_sessions(self, limit=20):
         """Prints a list of sessions using printer.table."""
         sessions = self._get_sessions()
         if not sessions:
             printer.info("No saved AI sessions found.")
             return
         
+        total = len(sessions)
+        if limit and total > limit:
+            sessions = sessions[:limit]
+            
         columns = ["ID", "Title", "Created At", "Model"]
         rows = [[s["id"], s["title"], s["created_at"], s["model"]] for s in sessions]
-        printer.table("AI Persisted Sessions", columns, rows)
+        
+        title = "AI Persisted Sessions"
+        if limit and total > limit:
+            title += f" (Showing last {limit} of {total})"
+            
+        printer.table(title, columns, rows)
+        if limit and total > limit:
+            printer.info(f"Use '--list --all' (if supported) or check the sessions directory to see all {total} sessions.")
 
     def load_session_data(self, session_id):
         """Loads a session's raw data by ID."""
@@ -917,8 +957,10 @@ class ai:
         return sessions[0]["id"] if sessions else None
 
     def _generate_session_id(self, query):
-        """Generates a unique session ID based on timestamp."""
+        """Generates a unique session ID based on timestamp and a random suffix."""
-        return datetime.datetime.now().strftime("%Y%m%d-%H%M%S")
+        ts = datetime.datetime.now().strftime("%Y%m%d-%H%M%S")
+        suffix = secrets.token_hex(2)
+        return f"{ts}-{suffix}"
 
     def save_session(self, history, title=None, model=None):
         """Saves current history to the session file."""
@@ -927,6 +969,8 @@ class ai:
             first_user_msg = next((m["content"] for m in history if m["role"] == "user"), "new-session")
             self.session_id = self._generate_session_id(first_user_msg)
             self.session_path = os.path.join(self.sessions_dir, f"{self.session_id}.json")
+        elif not self.session_path:
+            self.session_path = os.path.join(self.sessions_dir, f"{self.session_id}.json")
 
         # If it's a new file, we might want to set a better title
         if not os.path.exists(self.session_path) and not title:
@@ -964,16 +1008,22 @@ class ai:
 
     @MethodHook
     def ask(self, user_input, dryrun=False, chat_history=None, status=None, debug=False, stream=True, session_id=None, chunk_callback=None):
-        if not self.engineer_key:
+        is_engineer_keyless = "vertex" in self.engineer_model.lower() or "ollama" in self.engineer_model.lower() or "local" in self.engineer_model.lower()
-            raise ValueError("Engineer API key not configured. Use 'connpy config --engineer-api-key <key>' to set it.")
+        if not self.engineer_key and not self.engineer_auth and not is_engineer_keyless:
+            raise ValueError("Engineer API key or authentication not configured. Use 'connpy config --engineer-auth <auth>' to set it.")
             
         if chat_history is None: chat_history = []
         
         # Load session if provided and history is empty
-        if session_id and not chat_history:
+        if session_id:
-            session_data = self.load_session_data(session_id)
+            # Force the session_id even if it doesn't exist yet
-            if session_data:
+            self.session_id = session_id
-                chat_history = session_data.get("history", [])
+            self.session_path = os.path.join(self.sessions_dir, f"{session_id}.json")
+            
+            if not chat_history:
+                session_data = self.load_session_data(session_id)
+                if session_data:
+                    chat_history = session_data.get("history", [])
                 # If we loaded history, the caller might need it back
                 # But typically ask() is called in a loop with an external history object
 
@@ -1009,6 +1059,7 @@ class ai:
         tools = self._get_architect_tools() if current_brain == "architect" else self._get_engineer_tools()
         model = self.architect_model if current_brain == "architect" else self.engineer_model
         key = self.architect_key if current_brain == "architect" else self.engineer_key
+        current_auth = self.architect_auth if current_brain == "architect" else self.engineer_auth
 
         # Estructura optimizada para Prompt Caching (Solo para Anthropic directo, Vertex tiene reglas distintas)
         if "claude" in model.lower() and "vertex" not in model.lower():
@@ -1058,8 +1109,8 @@ class ai:
                 
                 label = "[architect][bold]Architect[/bold][/architect]" if current_brain == "architect" else "[engineer][bold]Engineer[/bold][/engineer]"
                 if status: 
-                    # Notify responder identity ONLY for web/remote clients (StatusBridge has is_web)
+                    # Notify responder identity for web/remote clients
-                    if getattr(status, "is_web", False):
+                    if getattr(status, "is_web", False) or getattr(status, "is_remote", False):
                         status.update(f"__RESPONDER__:{current_brain}")
                     status.update(f"{label} is thinking... (step {iteration})")
                 
@@ -1068,12 +1119,12 @@ class ai:
                     safe_messages = self._sanitize_messages(messages)
                     if stream:
                         response, streamed_response = self._stream_completion(
-                            model=model, messages=safe_messages, tools=tools, api_key=key,
+                            model=model, messages=safe_messages, tools=tools, auth=current_auth,
                             status=status, label=label, debug=debug, num_retries=3,
                             chunk_callback=chunk_callback
                         )
                     else:
-                        response = completion(model=model, messages=safe_messages, tools=tools, api_key=key, num_retries=3)
+                        response = completion(model=model, messages=safe_messages, tools=tools, num_retries=3, **current_auth)
                 except Exception as e:
                     if current_brain == "architect":
                         if status: status.update("[unavailable]Architect unavailable! Falling back to Engineer...")
@@ -1082,6 +1133,7 @@ class ai:
                         model = self.engineer_model
                         tools = self._get_engineer_tools()
                         key = self.engineer_key
+                        current_auth = self.engineer_auth
                         # Rebuild messages with Engineer system prompt and original user request
                         messages = [{"role": "system", "content": self.engineer_system_prompt}]
                         # Add chat history if exists (excluding system prompt)
@@ -1174,6 +1226,7 @@ class ai:
                         model = self.architect_model
                         tools = self._get_architect_tools()
                         key = self.architect_key
+                        current_auth = self.architect_auth
                         messages[0] = {"role": "system", "content": self.architect_system_prompt}
                         # Prepare handover context to inject AFTER all tool responses
                         handover_msg = f"HANDOVER FROM EXECUTION ENGINE\n\nReason: {args['reason']}\n\nContext: {args['context']}\n\nYou are now in control of this conversation."
@@ -1195,6 +1248,7 @@ class ai:
                         model = self.engineer_model
                         tools = self._get_engineer_tools()
                         key = self.engineer_key
+                        current_auth = self.engineer_auth
                         messages[0] = {"role": "system", "content": self.engineer_system_prompt}
                         # Prepare handover context to inject AFTER all tool responses
                         handover_msg = f"HANDOVER FROM ARCHITECT\n\nSummary: {args['summary']}\n\nYou are now back in control. Continue handling the user's requests."
@@ -1236,7 +1290,7 @@ class ai:
                     messages.append({"role": "user", "content": "Hard iteration limit reached. Please provide a summary of your findings so far."})
                     try:
                         safe_messages = self._sanitize_messages(messages)
-                        response = completion(model=model, messages=safe_messages, tools=[], api_key=key)
+                        response = completion(model=model, messages=safe_messages, tools=[], **current_auth)
                         resp_msg = response.choices[0].message
                         messages.append(resp_msg.model_dump(exclude_none=True))
                     except Exception as e:
@@ -1256,7 +1310,7 @@ class ai:
             try:
                 safe_messages = self._sanitize_messages(summary_messages)
                 # Use tools=None to force a text summary during interruption
-                response = completion(model=model, messages=safe_messages, tools=None, api_key=key)
+                response = completion(model=model, messages=safe_messages, tools=None, **current_auth)
                 resp_msg = response.choices[0].message
                 messages.append(resp_msg.model_dump(exclude_none=True))
                 
@@ -1393,6 +1447,7 @@ Node: {node_name}"""
         # Use models based on persona
         current_model = self.architect_model if persona == "architect" else self.engineer_model
         current_key = self.architect_key if persona == "architect" else self.engineer_key
+        current_auth = self.architect_auth if persona == "architect" else self.engineer_auth
 
         try:
             while iteration < max_iterations:
@@ -1402,8 +1457,8 @@ Node: {node_name}"""
                     model=current_model,
                     messages=messages,
                     tools=mcp_tools if mcp_tools else None,
-                    api_key=current_key,
+                    stream=True,
-                    stream=True
+                    **current_auth
                 )
                 
                 full_content = ""
@@ -1476,8 +1531,8 @@ Node: {node_name}"""
                     model=self.engineer_model,
                     messages=messages,
                     tools=None,
-                    api_key=self.engineer_key,
+                    stream=True,
-                    stream=True
+                    **self.engineer_auth
                 )
                 
                 full_content = ""

connpy/cli/ai_handler.py

@@ -15,13 +15,22 @@ class AIHandler:
 
     def dispatch(self, args):
         if args.list_sessions:
-            sessions = self.app.services.ai.list_sessions()
+            limit = 20 if not getattr(args, "all", False) else None
+            sessions, total = self.app.services.ai.list_sessions(limit=limit)
             if not sessions:
                 printer.info("No saved AI sessions found.")
                 return
+            
             columns = ["ID", "Title", "Created At", "Model"]
             rows = [[s["id"], s["title"], s["created_at"], s["model"]] for s in sessions]
-            printer.table("AI Persisted Sessions", columns, rows)
+            
+            title = "AI Persisted Sessions"
+            if limit and total > limit:
+                title += f" (Showing last {limit} of {total})"
+                
+            printer.table(title, columns, rows)
+            if limit and total > limit:
+                printer.info(f"Use '--list --all' to see all {total} sessions.")
             return
             
         if args.delete_session:
@@ -38,7 +47,7 @@ class AIHandler:
         # Determinar session_id para retomar
         session_id = None
         if args.resume:
-            sessions = self.app.services.ai.list_sessions()
+            sessions, _ = self.app.services.ai.list_sessions()
             session_id = sessions[0]["id"] if sessions else None
             if not session_id:
                 printer.warning("No previous session found to resume.")
@@ -57,15 +66,22 @@ class AIHandler:
             elif settings.get(key):
                 arguments[key] = settings.get(key)
 
+        for key in ["engineer_auth", "architect_auth"]:
+            cli_val = getattr(args, key, None)
+            if cli_val:
+                arguments[key] = self._parse_auth_value(cli_val[0])
+            elif settings.get(key):
+                arguments[key] = settings.get(key)
+        
         # Check keys only if running in local mode (not remote)
         if getattr(self.app.services, "mode", "local") == "local":
-            if not arguments.get("engineer_api_key"):
+            if not arguments.get("engineer_api_key") and not arguments.get("engineer_auth"):
-                printer.error("Engineer API key not configured. The chat cannot start.")
+                printer.error("Engineer API key/auth not configured. The chat cannot start.")
-                printer.info("Use 'connpy config --engineer-api-key <key>' to set it.")
+                printer.info("Use 'connpy config --engineer-api-key <key>' or 'connpy config --engineer-auth <auth>' to set it.")
                 sys.exit(1)
-            if not arguments.get("architect_api_key"):
+            if not arguments.get("architect_api_key") and not arguments.get("architect_auth"):
-                printer.warning("Architect API key not configured. Architect will be unavailable.")
+                printer.warning("Architect API key/auth not configured. Architect will be unavailable.")
-                printer.info("Use 'connpy config --architect-api-key <key>' to enable it.")
+                printer.info("Use 'connpy config --architect-api-key <key>' or 'connpy config --architect-auth <auth>' to enable it.")
 
         # El resto de la interacción el CLI la maneja con el agente subyacente
         self.app.myai = self.app.services.ai
@@ -102,7 +118,7 @@ class AIHandler:
                 if history:
                     mdprint(f"[debug]Analyzing {len(history)} previous messages...[/debug]\n")
             else:
-                printer.error(f"Could not load session {session_id}. Starting clean.")
+                printer.info(f"Session '{session_id}' not found. Starting clean.")
         
         if not history:
             mdprint(Rule(style="engineer"))
@@ -116,7 +132,7 @@ class AIHandler:
                 if user_query.lower() in ['exit', 'quit', 'bye', 'cancel']: break
                 
                 with console.status("[ai_status]Agent is thinking...") as status:
-                    result = self.app.myai.ask(user_query, chat_history=history, status=status, debug=args.debug, trust=args.trust, **self.ai_overrides)
+                    result = self.app.myai.ask(user_query, chat_history=history, status=status, debug=args.debug, trust=args.trust, session_id=session_id, **self.ai_overrides)
                 
                 new_history = result.get("chat_history")
                 if new_history is not None:
@@ -147,8 +163,7 @@ class AIHandler:
             action = mcp_args[0].lower()
             
             if action == "list":
-                settings = self.app.services.config_svc.get_settings()
+                mcp_servers = self.app.services.ai.list_mcp_servers()
-                mcp_servers = settings.get("ai", {}).get("mcp_servers", {})
                 if not mcp_servers:
                     printer.info("No MCP servers configured.")
                 else:
@@ -213,8 +228,7 @@ class AIHandler:
             from .forms import Forms
             self.app.cli_forms = Forms(self.app)
             
-        settings = self.app.services.config_svc.get_settings()
+        mcp_servers = self.app.services.ai.list_mcp_servers()
-        mcp_servers = settings.get("ai", {}).get("mcp_servers", {})
         
         result = self.app.cli_forms.mcp_wizard(mcp_servers)
         if not result:
@@ -249,3 +263,33 @@ class AIHandler:
                 
         except Exception as e:
             printer.error(str(e))
+
+    def _parse_auth_value(self, value):
+        if not value or value.lower() in ["none", "clear"]:
+            return None
+        import os
+        import yaml
+        import json
+        if os.path.exists(value):
+            try:
+                with open(value, "r") as f:
+                    content = f.read()
+                try:
+                    return json.loads(content)
+                except ValueError:
+                    return yaml.safe_load(content)
+            except Exception as e:
+                printer.error(f"Failed to read/parse auth file '{value}': {e}")
+                sys.exit(1)
+        
+        try:
+            return json.loads(value)
+        except ValueError:
+            try:
+                parsed = yaml.safe_load(value)
+                if isinstance(parsed, dict):
+                    return parsed
+                raise ValueError()
+            except Exception:
+                printer.error("Auth parameter must be a valid JSON/YAML string, or a path to a JSON/YAML file.")
+                sys.exit(1)

connpy/cli/config_handler.py

@@ -19,8 +19,10 @@ class ConfigHandler:
             "theme": self.set_theme,
             "engineer_model": self.set_ai_config,
             "engineer_api_key": self.set_ai_config,
+            "engineer_auth": self.set_ai_config,
             "architect_model": self.set_ai_config,
             "architect_api_key": self.set_ai_config,
+            "architect_auth": self.set_ai_config,
             "trusted_commands": self.set_ai_config,
             "service_mode": self.set_service_mode,
             "remote_host": self.set_remote_host,
@@ -127,9 +129,57 @@ class ConfigHandler:
         try:
             settings = self.app.services.config_svc.get_settings()
             aiconfig = settings.get("ai", {})
-            aiconfig[args.command] = args.data[0]
+            val = args.data[0]
+            
+            # Check for unset/clear request
+            if val.lower() in ["none", "clear", ""]:
+                if args.command in aiconfig:
+                    del aiconfig[args.command]
+            else:
+                # If configuring auth, parse as dictionary (JSON/YAML or file path)
+                if args.command in ["engineer_auth", "architect_auth"]:
+                    parsed_val = self._parse_auth_value(val)
+                    if parsed_val is not None:
+                        aiconfig[args.command] = parsed_val
+                    else:
+                        if args.command in aiconfig:
+                            del aiconfig[args.command]
+                else:
+                    aiconfig[args.command] = val
+            
             self.app.services.config_svc.update_setting("ai", aiconfig)
             printer.success("Config saved")
-        except ConnpyError as e:
+        except (ConnpyError, InvalidConfigurationError) as e:
             printer.error(str(e))
 
+    def _parse_auth_value(self, value):
+        if value.lower() in ["none", "clear", ""]:
+            return None
+        
+        # Check if it's a file path
+        import os
+        if os.path.exists(value):
+            try:
+                with open(value, "r") as f:
+                    content = f.read()
+                import json
+                try:
+                    return json.loads(content)
+                except ValueError:
+                    return yaml.safe_load(content)
+            except Exception as e:
+                raise InvalidConfigurationError(f"Failed to read/parse auth file '{value}': {e}")
+        
+        # Try parsing as inline JSON/YAML
+        try:
+            import json
+            return json.loads(value)
+        except ValueError:
+            try:
+                parsed = yaml.safe_load(value)
+                if isinstance(parsed, dict):
+                    return parsed
+                raise ValueError()
+            except Exception:
+                raise InvalidConfigurationError("Auth parameter must be a valid JSON/YAML string, or a path to a JSON/YAML file.")
+

connpy/cli/helpers.py

@@ -1,10 +1,81 @@
 import os
 import inquirer
+from inquirer.themes import Default, term
+
 try:
     from pyfzf.pyfzf import FzfPrompt
 except ImportError:
     FzfPrompt = None
 
+def hex_to_blessed(hex_str):
+    """Convert hex color string to blessed/ansi format."""
+    if not hex_str or not isinstance(hex_str, str):
+        return term.normal
+    
+    # Check for bold prefix
+    prefix = ""
+    if hex_str.startswith('bold '):
+        prefix = term.bold
+        hex_str = hex_str.replace('bold ', '').strip()
+    
+    # If it's a standard color name
+    if not hex_str.startswith('#'):
+        return prefix + getattr(term, hex_str, term.normal)
+    
+    # Parse hex
+    try:
+        h = hex_str.lstrip('#')
+        if len(h) == 3:
+            h = ''.join([c*2 for c in h])
+        r = int(h[0:2], 16)
+        g = int(h[2:4], 16)
+        b = int(h[4:6], 16)
+        
+        # Try RGB, fallback to standard cyan if it fails or returns empty
+        try:
+            c = term.color_rgb(r, g, b)
+            if not c: # Some terms return empty for RGB
+                return prefix + term.cyan
+            return prefix + c
+        except:
+            return prefix + term.cyan
+    except:
+        return prefix + term.normal
+
+# Custom inquirer theme matching connpy colors
+class ConnpyTheme(Default):
+    def __init__(self):
+        super().__init__()
+        try:
+            from ..printer import _global_active_styles
+            # Use user_prompt as primary accent, fallback to info/cyan
+            accent = _global_active_styles.get("user_prompt", _global_active_styles.get("info", "cyan"))
+            accent_color = hex_to_blessed(accent)
+            
+            self.Question.mark_color = accent_color
+            self.List.selection_color = accent_color
+            self.List.selection_cursor = ">"
+        except:
+            # Absolute fallback to standard cyan
+            self.Question.mark_color = term.cyan
+            self.List.selection_color = term.bold_cyan
+            self.List.selection_cursor = ">"
+
+def get_theme():
+    """Returns a fresh instance of the theme with current colors."""
+    return ConnpyTheme()
+
+class ThemeProxy:
+    """Proxy to ensure theme colors are resolved at runtime."""
+    def __getattr__(self, name):
+        return getattr(get_theme(), name)
+    def __iter__(self):
+        return iter(get_theme())
+    def __getitem__(self, item):
+        return get_theme()[item]
+
+theme = ThemeProxy()
+
 def get_config_dir():
     home = os.path.expanduser("~")
     defaultdir = os.path.join(home, '.config/conn')
@@ -56,7 +127,7 @@ def choose(app, list_, name, action):
             return answer[0]
     else:
         questions = [inquirer.List(name, message="Pick {} to {}:".format(name,action), choices=list_, carousel=True)]
-        answer = inquirer.prompt(questions)
+        answer = inquirer.prompt(questions, theme=theme)
         if answer == None:
             return None
         else:

connpy/cli/terminal_ui.py

@@ -134,7 +134,8 @@ class CopilotInterface:
                 if state['context_mode'] == self.mode_single:
                     active_raw = raw_bytes[start:end]
                 else:
-                    active_raw = raw_bytes[start:]
+                    # Concat only the bytes of valid blocks to skip intermediate empty/cancelled prompt noise
+                    active_raw = b"".join(raw_bytes[b[0]:b[1]] for b in blocks[idx:])
                 return preview + "\n" + log_cleaner(active_raw.decode(errors='replace'))
 
             def get_prompt_text():
@@ -335,6 +336,7 @@ class CopilotInterface:
                 persona_title = "Network Architect" if active_persona == "architect" else "Network Engineer"
                 
                 active_buffer = get_active_buffer()
+                
                 live_text = ""
                 first_chunk = True
                 

connpy/completion.py

@@ -181,11 +181,28 @@ def _build_tree(nodes, folders, profiles, plugins, configdir):
     ai_dict = {"__exclude_used__": True, "--help": None, "-h": None}
     for opt in ["--engineer-model", "--engineer-api-key", "--architect-model", "--architect-api-key"]:
         ai_dict[opt] = {"*": ai_dict} # takes value, loops back
+    ai_dict["--engineer-auth"] = {"__extra__": lambda w: get_cwd(w, "--engineer-auth"), "*": ai_dict}
+    ai_dict["--architect-auth"] = {"__extra__": lambda w: get_cwd(w, "--architect-auth"), "*": ai_dict}
     for opt in ["--debug", "--trust", "--list", "--list-sessions", "--session", "--resume", "--delete", "--delete-session", "-y"]:
         ai_dict[opt] = ai_dict # takes no value, loops back
     ai_dict["--mcp"] = mcp_dict
     ai_dict["*"] = ai_dict
 
+    config_dict = {
+        "--allow-uppercase": ["true", "false"],
+        "--fzf": ["true", "false"],
+        "--completion": ["bash", "zsh"],
+        "--fzf-wrapper": ["bash", "zsh"],
+        "--service-mode": ["local", "remote"],
+        "--sync-remote": ["true", "false"],
+        "--help": None, "-h": None,
+    }
+    for opt in ["--keepalive", "--engineer-model", "--engineer-api-key", "--architect-model", "--architect-api-key", "--theme", "--remote", "--trusted-commands"]:
+        config_dict[opt] = {"*": config_dict}
+    config_dict["--configfolder"] = {"__extra__": lambda w: get_cwd(w, "--configfolder", True), "*": config_dict}
+    config_dict["--engineer-auth"] = {"__extra__": lambda w: get_cwd(w, "--engineer-auth"), "*": config_dict}
+    config_dict["--architect-auth"] = {"__extra__": lambda w: get_cwd(w, "--architect-auth"), "*": config_dict}
+
     mv_state = {"__extra__": _nodes, "--help": None, "-h": None}
     cp_state = {"__extra__": _nodes, "--help": None, "-h": None}
     ls_state = {
@@ -280,22 +297,7 @@ def _build_tree(nodes, folders, profiles, plugins, configdir):
             "--list": None, "--help": None,
             "-h": None,
         },
-        "config": {
+        "config": config_dict,
-            "--allow-uppercase": ["true", "false"],
-            "--fzf": ["true", "false"],
-            "--keepalive": None,
-            "--completion": ["bash", "zsh"],
-            "--fzf-wrapper": ["bash", "zsh"],
-            "--configfolder": lambda w: get_cwd(w, "--configfolder", True),
-            "--engineer-model": None, "--engineer-api-key": None,
-            "--architect-model": None, "--architect-api-key": None,
-            "--theme": None,
-            "--service-mode": ["local", "remote"],
-            "--remote": None,
-            "--sync-remote": ["true", "false"],
-            "--trusted-commands": None,
-            "--help": None, "-h": None,
-        },
         "sync": {
             "--login": None, "--logout": None,
             "--status": None, "--list": None,

connpy/connapp.py

@@ -276,11 +276,14 @@ class connapp:
         aiparser.add_argument("ask", nargs='*', help="Ask connpy AI something")
         aiparser.add_argument("--engineer-model", nargs=1, help="Override engineer model")
         aiparser.add_argument("--engineer-api-key", nargs=1, help="Override engineer api key")
+        aiparser.add_argument("--engineer-auth", nargs=1, help="Override engineer auth (inline JSON/YAML or file path)")
         aiparser.add_argument("--architect-model", nargs=1, help="Override architect model")
         aiparser.add_argument("--architect-api-key", nargs=1, help="Override architect api key")
+        aiparser.add_argument("--architect-auth", nargs=1, help="Override architect auth (inline JSON/YAML or file path)")
         aiparser.add_argument("--debug", action="store_true", help="Show AI reasoning and tool calls")
         aiparser.add_argument("-y", "--trust", action="store_true", help="Trust AI to execute unsafe commands without confirmation")
         aiparser.add_argument("--list", "--list-sessions", dest="list_sessions", action="store_true", help="List saved AI sessions")
+        aiparser.add_argument("--all", action="store_true", help="Show all sessions without limit")
         aiparser.add_argument("--session", nargs=1, help="Resume a specific AI session by ID")
         aiparser.add_argument("--resume", action="store_true", help="Resume the most recent AI session")
         aiparser.add_argument("--delete", "--delete-session", dest="delete_session", nargs=1, help="Delete an AI session by ID")
@@ -340,11 +343,13 @@ class connapp:
         configcrud.add_argument("--configfolder", dest="configfolder", nargs=1, action=self._store_type, help="Set the default location for config file", metavar="FOLDER")
         configcrud.add_argument("--engineer-model", dest="engineer_model", nargs=1, action=self._store_type, help="Set engineer model", metavar="MODEL")
         configcrud.add_argument("--engineer-api-key", dest="engineer_api_key", nargs=1, action=self._store_type, help="Set engineer api_key", metavar="API_KEY")
+        configcrud.add_argument("--engineer-auth", dest="engineer_auth", nargs=1, action=self._store_type, help="Set engineer auth (inline JSON/YAML or file path)", metavar="AUTH")
         configcrud.add_argument("--theme", dest="theme", nargs=1, action=self._store_type, help="Set application theme (dark, light, or YAML file path)", metavar="THEME")
         configcrud.add_argument("--service-mode", dest="service_mode", nargs=1, action=self._store_type, help="Set the backend service mode (local or remote)", choices=["local", "remote"])
         configcrud.add_argument("--remote", dest="remote_host", nargs=1, action=self._store_type, help="Connect to a remote connpy service via gRPC", metavar="HOST:PORT")
         configcrud.add_argument("--architect-model", dest="architect_model", nargs=1, action=self._store_type, help="Set architect model", metavar="MODEL")
         configcrud.add_argument("--architect-api-key", dest="architect_api_key", nargs=1, action=self._store_type, help="Set architect api_key", metavar="API_KEY")
+        configcrud.add_argument("--architect-auth", dest="architect_auth", nargs=1, action=self._store_type, help="Set architect auth (inline JSON/YAML or file path)", metavar="AUTH")
         configcrud.add_argument("--sync-remote", dest="sync_remote", nargs=1, action=self._store_type, help="Sync remote nodes to Google Drive", choices=["true","false"])
         configparser.add_argument("--trusted-commands", dest="trusted_commands", nargs=1, action=self._store_type, help="Set custom trusted commands regexes (comma separated)", metavar="REGEX,REGEX")
         configparser.set_defaults(func=self._config.dispatch)

connpy/core.py

@@ -315,8 +315,11 @@ class node:
 
 
     def _setup_interact_environment(self, debug=False, logger=None, async_mode=False):
-        size = re.search('columns=([0-9]+).*lines=([0-9]+)',str(os.get_terminal_size()))
+        try:
-        self.child.setwinsize(int(size.group(2)),int(size.group(1)))
+            size = re.search('columns=([0-9]+).*lines=([0-9]+)',str(os.get_terminal_size()))
+            self.child.setwinsize(int(size.group(2)),int(size.group(1)))
+        except OSError:
+            pass
         if logger:
             port_str = f":{self.port}" if self.port and self.protocol not in ["ssm", "kubectl", "docker"] else ""
             logger("success", f"Connected to {self.unique} at {self.host}{port_str} via: {self.protocol}")
@@ -353,6 +356,7 @@ class node:
 
     async def _async_interact_loop(self, local_stream, resize_callback, copilot_handler=None):
         local_stream.setup(resize_callback=resize_callback)
+        self.current_local_stream = local_stream
         try:
             child_fd = self.child.child_fd
             
@@ -435,11 +439,19 @@ class node:
                     # Remove any stray \x00 bytes and forward normally
                     clean_data = data.replace(b'\x00', b'')
                     if clean_data:
-                        # Track command boundaries when user hits Enter
+                        # Track command boundaries when user hits Enter or presses Ctrl+C
-                        if hasattr(self, 'mylog') and (b'\r' in clean_data or b'\n' in clean_data):
+                        if hasattr(self, 'mylog') and (b'\r' in clean_data or b'\n' in clean_data or b'\x03' in clean_data):
-                            self.cmd_byte_positions.append((self.mylog.tell(), None))
+                            pos = self.mylog.tell()
+                            marker_cmd = "CANCELLED" if b'\x03' in clean_data else None
+                            self.cmd_byte_positions.append((pos, marker_cmd))
+                            if hasattr(self, 'current_local_stream') and self.current_local_stream is not None:
+                                try:
+                                    await self.current_local_stream.write(f'\x1b]133;B;{pos}\x07'.encode())
+                                except Exception:
+                                    pass
 
-                        try:                            os.write(child_fd, clean_data)
+                        try:
+                            os.write(child_fd, clean_data)
                         except OSError:
                             break
                         self.lastinput = time()
@@ -559,6 +571,7 @@ class node:
                 except Exception:
                     pass
         finally:
+            self.current_local_stream = None
             local_stream.teardown()
 
     @MethodHook
@@ -587,6 +600,11 @@ class node:
             if cmd != slc and hasattr(self, 'cmd_byte_positions') and self.cmd_byte_positions is not None:
                 log_pos = self.mylog.tell() if hasattr(self, 'mylog') else 0
                 self.cmd_byte_positions.append((log_pos, cmd))
+                if hasattr(self, 'current_local_stream') and self.current_local_stream is not None:
+                    try:
+                        await self.current_local_stream.write(f'\x1b]133;B;{log_pos}\x07'.encode())
+                    except Exception:
+                        pass
             
             # Write physically to PTY
             os.write(child_fd, (cmd + "\n").encode())

connpy/grpc_layer/server.py

@@ -375,7 +375,9 @@ class NodeServicer(connpy_pb2_grpc.NodeServiceServicer):
 
                 asyncio.run(n._async_interact_loop(remote_stream, resize_callback, copilot_handler=remote_copilot_handler))
             except Exception as e:
-                pass
+                import traceback
+                print(f"[ERROR in run_async_loop] {e}")
+                traceback.print_exc()
             finally:
                 n._teardown_interact_environment()
                 response_queue.put(None)  # Signal EOF
@@ -481,6 +483,7 @@ class ProfileServicer(connpy_pb2_grpc.ProfileServiceServicer):
         self.service = ProfileService(config)
         self.node_service = NodeService(config)
 
+
     @handle_errors
     def list_profiles(self, request, context):
         f = request.filter_str if request.filter_str else None
@@ -729,6 +732,7 @@ class StatusBridge:
         self.on_interrupt = self._force_interrupt
         self.thread = None
         self.is_web = is_web
+        self.is_remote = True
 
     def _force_interrupt(self):
         """Forcefully raise KeyboardInterrupt in the target thread."""
@@ -860,9 +864,11 @@ class AIServicer(connpy_pb2_grpc.AIServiceServicer):
                 print(f"AI Task Error: {e}")
                 traceback.print_exc()
                 chunk_queue.put(("status", f"Error: {str(e)}"))
+                # Crucial: always send final_mark to avoid client deadlock
+                chunk_queue.put(("final_mark", {"response": f"Error: {str(e)}", "chat_history": history, "error": True}))
 
         def request_listener():
-            nonlocal bridge, is_web, ai_thread, agent_instance
+            nonlocal bridge, is_web, ai_thread, agent_instance, history
             try:
                 for req in request_iterator:
                     if req.interrupt:
@@ -876,12 +882,21 @@ class AIServicer(connpy_pb2_grpc.AIServiceServicer):
 
                     if req.input_text:
                         is_web = "web" in (req.session_id or "").lower() or (req.session_id or "").lower().startswith("ws-")
+                        
+                        # Hydrate history from client if it's the first interaction in this stream
+                        if not history and req.chat_history:
+                            from .utils import from_value
+                            history = from_value(req.chat_history) or []
                         if not bridge:
                             bridge = StatusBridge(chunk_queue, request_queue=request_queue, is_web=is_web)
 
                         overrides = {}
                         if req.engineer_model: overrides["engineer_model"] = req.engineer_model
                         if req.engineer_api_key: overrides["engineer_api_key"] = req.engineer_api_key
+                        if req.architect_model: overrides["architect_model"] = req.architect_model
+                        if req.architect_api_key: overrides["architect_api_key"] = req.architect_api_key
+                        if req.HasField("engineer_auth"): overrides["engineer_auth"] = from_struct(req.engineer_auth)
+                        if req.HasField("architect_auth"): overrides["architect_auth"] = from_struct(req.architect_auth)
                         
                         # Start AI in its own thread so we can keep listening for interrupts
                         ai_thread = threading.Thread(
@@ -946,7 +961,8 @@ class AIServicer(connpy_pb2_grpc.AIServiceServicer):
 
     @handle_errors
     def list_sessions(self, request, context):
-        return connpy_pb2.ValueResponse(data=to_value(self.service.list_sessions()))
+        sessions, total = self.service.list_sessions()
+        return connpy_pb2.ValueResponse(data=to_value(sessions))
 
     @handle_errors
     def delete_session(self, request, context):
@@ -955,7 +971,8 @@ class AIServicer(connpy_pb2_grpc.AIServiceServicer):
 
     @handle_errors
     def configure_provider(self, request, context):
-        self.service.configure_provider(request.provider, request.model, request.api_key)
+        auth_dict = from_struct(request.auth) if request.HasField("auth") else None
+        self.service.configure_provider(request.provider, request.model, request.api_key, auth=auth_dict)
         return Empty()
     
     @handle_errors
@@ -969,6 +986,11 @@ class AIServicer(connpy_pb2_grpc.AIServiceServicer):
         )
         return Empty()
 
+    @handle_errors
+    def list_mcp_servers(self, request, context):
+        mcp_servers = self.service.list_mcp_servers()
+        return connpy_pb2.ValueResponse(data=to_value(mcp_servers))
+
     @handle_errors
     def load_session_data(self, request, context):
         return connpy_pb2.StructResponse(data=to_struct(self.service.load_session_data(request.value)))

connpy/grpc_layer/stubs.py

@@ -745,6 +745,10 @@ class AIStub:
         )
         if chat_history is not None:
             initial_req.chat_history.CopyFrom(to_value(chat_history))
+        if "engineer_auth" in overrides and overrides["engineer_auth"]:
+            initial_req.engineer_auth.CopyFrom(to_struct(overrides["engineer_auth"]))
+        if "architect_auth" in overrides and overrides["architect_auth"]:
+            initial_req.architect_auth.CopyFrom(to_struct(overrides["architect_auth"]))
             
         req_queue.put(initial_req)
 
@@ -758,6 +762,7 @@ class AIStub:
         
         full_content = ""
         header_printed = False
+        current_responder = "engineer"
         final_result = {"response": "", "chat_history": []}
 
         # Background thread to pull responses from gRPC into a local queue
@@ -802,6 +807,10 @@ class AIStub:
                     break
 
                 if response.status_update:
+                    if response.status_update.startswith("__RESPONDER__:"):
+                        current_responder = response.status_update.split(":")[1].lower()
+                        continue
+                        
                     if response.requires_confirmation:
                         if status: status.stop()
                         
@@ -854,7 +863,9 @@ class AIStub:
                             stable_console = RichConsole(theme=connpy_theme, file=get_original_stdout())
                             
                             # Print header on first chunk
-                            stable_console.print(Rule("[bold engineer]Network Engineer[/bold engineer]", style="engineer"))
+                            alias = "architect" if current_responder == "architect" else "engineer"
+                            role_label = "Network Architect" if current_responder == "architect" else "Network Engineer"
+                            stable_console.print(Rule(f"[bold {alias}]{role_label}[/bold {alias}]", style=alias))
                             header_printed = True
                             
                             # Initialize parser
@@ -906,16 +917,23 @@ class AIStub:
         return self.stub.confirm(connpy_pb2.StringRequest(value=input_text)).value
 
     @handle_errors
-    def list_sessions(self):
+    def list_sessions(self, limit=None):
-        return from_value(self.stub.list_sessions(Empty()).data)
+        from .utils import from_value
+        res = self.stub.list_sessions(Empty())
+        sessions = from_value(res.data) or []
+        if limit and len(sessions) > limit:
+            return sessions[:limit], len(sessions)
+        return sessions, len(sessions)
 
     @handle_errors
     def delete_session(self, session_id):
         self.stub.delete_session(connpy_pb2.StringRequest(value=session_id))
 
     @handle_errors
-    def configure_provider(self, provider, model=None, api_key=None):
+    def configure_provider(self, provider, model=None, api_key=None, auth=None):
         req = connpy_pb2.ProviderRequest(provider=provider, model=model or "", api_key=api_key or "")
+        if auth:
+            req.auth.CopyFrom(to_struct(auth))
         self.stub.configure_provider(req)
 
     @handle_errors
@@ -929,6 +947,11 @@ class AIStub:
         )
         self.stub.configure_mcp(req)
 
+    @handle_errors
+    def list_mcp_servers(self):
+        res = self.stub.list_mcp_servers(Empty())
+        return from_value(res.data) or {}
+
     @handle_errors
     def load_session_data(self, session_id):
         return from_struct(self.stub.load_session_data(connpy_pb2.StringRequest(value=session_id)).data)

connpy/proto/connpy.proto

@@ -70,6 +70,7 @@ service AIService {
     rpc delete_session (StringRequest) returns (google.protobuf.Empty) {}
     rpc configure_provider (ProviderRequest) returns (google.protobuf.Empty) {}
     rpc configure_mcp (MCPRequest) returns (google.protobuf.Empty) {}
+    rpc list_mcp_servers (google.protobuf.Empty) returns (ValueResponse) {}
     rpc load_session_data (StringRequest) returns (StructResponse) {}
 }
 
@@ -234,6 +235,8 @@ message AskRequest {
     bool trust = 10;
     string confirmation_answer = 11;
     bool interrupt = 12;
+    google.protobuf.Struct engineer_auth = 13;
+    google.protobuf.Struct architect_auth = 14;
 }
 
 message AIResponse {
@@ -254,6 +257,7 @@ message ProviderRequest {
     string provider = 1;
     string model = 2;
     string api_key = 3;
+    google.protobuf.Struct auth = 4;
 }
 
 message IntRequest {

connpy/services/ai_service.py

@@ -6,6 +6,37 @@ from connpy.utils import log_cleaner
 class AIService(BaseService):
     """Business logic for interacting with AI agents and LLM configurations."""
 
+    def _clean_cisco_scrolling(self, text: str) -> str:
+        """Resolves horizontal scrolling artifacts (backspaces, \r, ANSI) by merging overlapping segments."""
+        def merge_overlapping(s1, s2):
+            s2_clean = s2.lstrip(' $')
+            max_overlap = min(len(s1), len(s2_clean))
+            for i in range(max_overlap, 0, -1):
+                if s1[-i:] == s2_clean[:i]:
+                    return s1 + s2_clean[i:]
+            return s1 + s2_clean
+
+        scroll_re = re.compile(r'(\x08{5,}\s*\$?|\$\r|\x1b\[\d+[GD]\s*\$?)')
+        parts = scroll_re.split(text)
+        merged = ""
+        
+        for part in parts:
+            if scroll_re.match(part):
+                continue
+                
+            cleaned = log_cleaner(part)
+            if not merged:
+                merged = cleaned
+            else:
+                merged_lines = merged.split('\n')
+                cleaned_lines = cleaned.split('\n')
+                
+                merged_lines[-1] = merge_overlapping(merged_lines[-1], cleaned_lines[0])
+                merged_lines.extend(cleaned_lines[1:])
+                merged = "\n".join(merged_lines)
+                
+        return merged
+
     def build_context_blocks(self, raw_bytes: bytes, cmd_byte_positions: list, node_info: dict, last_line: str = "") -> list:
         """Identifies command blocks in the terminal history."""
         blocks = []
@@ -27,28 +58,69 @@ class AIService(BaseService):
                 prev_pos = cmd_byte_positions[i-1][0]
                 
                 if known_cmd:
-                    prev_chunk = raw_bytes[prev_pos:pos]
+                    if known_cmd == "CANCELLED":
-                    prev_cleaned = log_cleaner(prev_chunk.decode(errors='replace'))
+                        parsed_positions.append({"pos": pos, "type": "CANCELLED", "preview": ""})
-                    prev_lines = [l for l in prev_cleaned.split('\n') if l.strip()]
+                    else:
-                    prompt_text = prev_lines[-1].strip() if prev_lines else ""
+                        prev_chunk = raw_bytes[prev_pos:pos]
-                    preview = f"{prompt_text}{known_cmd}" if prompt_text else known_cmd
+                        prev_cleaned = self._clean_cisco_scrolling(prev_chunk.decode(errors='replace'))
-                    parsed_positions.append({"pos": pos, "type": "VALID_CMD", "preview": preview[:80]})
+                        prev_lines = [l for l in prev_cleaned.split('\n') if l.strip()]
+                        prompt_text = prev_lines[-1].strip() if prev_lines else ""
+                        preview = f"{prompt_text}{known_cmd}" if prompt_text else known_cmd
+                        
+                        if len(preview) > 80:
+                            preview = preview[:77] + "..."
+                        parsed_positions.append({"pos": pos, "type": "VALID_CMD", "preview": preview})
                 else:
                     chunk = raw_bytes[prev_pos:pos]
-                    cleaned = log_cleaner(chunk.decode(errors='replace'))
-                    lines = [l for l in cleaned.split('\n') if l.strip()]
-                    preview = lines[-1].strip() if lines else ""
                     
-                    if preview:
+                    cleaned = self._clean_cisco_scrolling(chunk.decode(errors='replace'))
-                        match = prompt_re.search(preview)
+                    lines = [l for l in cleaned.split('\n') if l.strip()]
-                        if match:
+                    
-                            cmd_text = preview[match.end():].strip()
+                    found_in_pass1 = False
-                            if cmd_text:
+                    if lines:
-                                parsed_positions.append({"pos": pos, "type": "VALID_CMD", "preview": preview[:80]})
+                        # Search backwards through the last few lines for the prompt
-                            else:
+                        for idx in range(len(lines) - 1, max(-1, len(lines) - 10), -1):
-                                parsed_positions.append({"pos": pos, "type": "EMPTY_PROMPT", "preview": ""})
+                            match = prompt_re.search(lines[idx])
-                        else:
+                            if match:
-                            parsed_positions.append({"pos": pos, "type": "SCROLLING", "preview": ""})
+                                ptxt = match.group(0).strip()
+                                cmd_first_line = lines[idx][match.end():].strip()
+                                cmd_rest = [l.strip() for l in lines[idx+1:]]
+                                cmd_text = " ".join([cmd_first_line] + cmd_rest).strip()
+                                
+                                if cmd_text:
+                                    pv = f"{ptxt} {cmd_text}".strip()
+                                    if len(pv) > 80:
+                                        pv = pv[:77] + "..."
+                                    parsed_positions.append({"pos": pos, "type": "VALID_CMD", "preview": pv})
+                                else:
+                                    parsed_positions.append({"pos": pos, "type": "EMPTY_PROMPT", "preview": ""})
+                                found_in_pass1 = True
+                                break
+                        
+                        if not found_in_pass1:
+                            # Fallback: The prompt might have been isolated in the previous chunk 
+                            # due to asynchronous network delays splitting the output exactly at the newline.
+                            prev_was_valid_cmd = i >= 2 and parsed_positions[i-2]["type"] == "VALID_CMD"
+                            if prev_pos > 0 and not prev_was_valid_cmd:
+                                # Fetch the very last chunk that we just processed
+                                prev_prev_pos = cmd_byte_positions[i-2][0] if i >= 2 else 0
+                                prev_chunk_text = self._clean_cisco_scrolling(raw_bytes[prev_prev_pos:prev_pos].decode(errors='replace'))
+                                prev_lines_text = [l for l in prev_chunk_text.split('\n') if l.strip()]
+                                
+                                if prev_lines_text:
+                                    prev_match = prompt_re.search(prev_lines_text[-1])
+                                    if prev_match:
+                                        ptxt = prev_match.group(0).strip()
+                                        cmd_text = " ".join([l.strip() for l in lines]).strip()
+                                        if cmd_text:
+                                            pv = f"{ptxt} {cmd_text}".strip()
+                                            if len(pv) > 80:
+                                                pv = pv[:77] + "..."
+                                            parsed_positions.append({"pos": pos, "type": "VALID_CMD", "preview": pv})
+                                            found_in_pass1 = True
+                            
+                            if not found_in_pass1:
+                                parsed_positions.append({"pos": pos, "type": "SCROLLING", "preview": ""})
                     else:
                         parsed_positions.append({"pos": pos, "type": "SCROLLING", "preview": ""})
 
@@ -61,11 +133,11 @@ class AIService(BaseService):
                 start_pos = item["pos"]
                 preview = item["preview"]
                 
-                # Find the end position: next VALID_CMD or EMPTY_PROMPT
+                # Find the end position: next VALID_CMD or EMPTY_PROMPT or CANCELLED
                 end_pos = current_prompt_pos
                 for j in range(i + 1, len(parsed_positions)):
                     next_item = parsed_positions[j]
-                    if next_item["type"] in ("VALID_CMD", "EMPTY_PROMPT"):
+                    if next_item["type"] in ("VALID_CMD", "EMPTY_PROMPT", "CANCELLED"):
                         end_pos = next_item["pos"]
                         break
                 
@@ -167,11 +239,14 @@ class AIService(BaseService):
         return await asyncio.wrap_future(future)
 
 
-    def list_sessions(self):
+    def list_sessions(self, limit=None):
-        """Return a list of all saved AI sessions."""
+        """Return a list of saved AI sessions, optionally limited."""
         from connpy.ai import ai
         agent = ai(self.config)
-        return agent._get_sessions()
+        sessions = agent._get_sessions()
+        if limit and len(sessions) > limit:
+            return sessions[:limit], len(sessions)
+        return sessions, len(sessions)
 
     def delete_session(self, session_id):
         """Delete an AI session by ID."""
@@ -183,13 +258,15 @@ class AIService(BaseService):
         else:
             raise InvalidConfigurationError(f"Session '{session_id}' not found.")
 
-    def configure_provider(self, provider, model=None, api_key=None):
+    def configure_provider(self, provider, model=None, api_key=None, auth=None):
         """Update AI provider settings in the configuration."""
         settings = self.config.config.get("ai", {})
         if model:
             settings[f"{provider}_model"] = model
         if api_key:
             settings[f"{provider}_api_key"] = api_key
+        if auth is not None:
+            settings[f"{provider}_auth"] = auth
             
         self.config.config["ai"] = settings
         self.config._saveconfig(self.config.file)
@@ -228,6 +305,11 @@ class AIService(BaseService):
         self.config.config["ai"] = ai_settings
         self.config._saveconfig(self.config.file)
 
+    def list_mcp_servers(self) -> dict:
+        """Get the configured MCP servers."""
+        ai_settings = self.config.config.get("ai", {})
+        return ai_settings.get("mcp_servers", {})
+
     def load_session_data(self, session_id):
         """Load a session's raw data by ID."""
         from connpy.ai import ai

connpy/tests/test_ai.py

@@ -23,7 +23,7 @@ class TestAIInit:
         myai = ai(config)
         with pytest.raises(ValueError) as exc:
             myai.ask("hello")
-        assert "Engineer API key not configured" in str(exc.value)
+        assert "Engineer API key or authentication not configured" in str(exc.value)
 
     def test_init_missing_architect_key_warns(self, ai_config, capsys, mock_litellm):
         """Warns if architect key is missing but doesn't crash."""
@@ -58,6 +58,77 @@ class TestAIInit:
                     pass  # May fail on other file opens, that's ok
 
 
+# =========================================================================
+# AI Auth Dict tests
+# =========================================================================
+
+class TestAIAuthDict:
+    def test_init_with_auth_dict(self, ai_config):
+        """Initializes correctly when auth dicts are configured."""
+        from connpy.ai import ai
+        ai_config.config["ai"]["engineer_api_key"] = None
+        ai_config.config["ai"]["architect_api_key"] = None
+        ai_config.config["ai"]["engineer_auth"] = {"my_key": "my_val"}
+        ai_config.config["ai"]["architect_auth"] = {"another_key": "another_val"}
+        myai = ai(ai_config)
+        assert myai.engineer_auth == {"my_key": "my_val"}
+        assert myai.architect_auth == {"another_key": "another_val"}
+
+    def test_compat_key_injection(self, ai_config):
+        """Injects API key into auth dict if auth is empty or doesn't have it."""
+        from connpy.ai import ai
+        ai_config.config["ai"]["engineer_api_key"] = "compat-eng-key"
+        ai_config.config["ai"]["architect_api_key"] = "compat-arch-key"
+        ai_config.config["ai"]["engineer_auth"] = {}
+        ai_config.config["ai"]["architect_auth"] = {}
+        myai = ai(ai_config)
+        assert myai.engineer_auth == {"api_key": "compat-eng-key"}
+        assert myai.architect_auth == {"api_key": "compat-arch-key"}
+
+    def test_has_architect_keyless(self, ai_config):
+        """Evaluates has_architect correctly for keyless models and auth configs."""
+        from connpy.ai import ai
+        # 1. Keyless model (Vertex)
+        ai_config.config["ai"]["architect_api_key"] = None
+        ai_config.config["ai"]["architect_auth"] = {}
+        ai_config.config["ai"]["architect_model"] = "vertex/gemini-pro"
+        myai = ai(ai_config)
+        assert myai.has_architect is True
+
+        # 2. Architect auth dict is set
+        ai_config.config["ai"]["architect_model"] = "custom-model"
+        ai_config.config["ai"]["architect_auth"] = {"vertex_project": "proj-1"}
+        myai = ai(ai_config)
+        assert myai.has_architect is True
+
+    def test_ask_unpacks_auth_dict(self, ai_config, mock_litellm):
+        """Verifies that ask unpacks engineer_auth when calling completion."""
+        from connpy.ai import ai
+        ai_config.config["ai"]["engineer_api_key"] = None
+        ai_config.config["ai"]["engineer_auth"] = {"vertex_project": "my-project", "vertex_location": "us-east1"}
+        myai = ai(ai_config)
+        myai.ask("test query", stream=False)
+        # Check mock_litellm completion call
+        mock_litellm["completion"].assert_called()
+        kwargs = mock_litellm["completion"].call_args.kwargs
+        assert kwargs.get("vertex_project") == "my-project"
+        assert kwargs.get("vertex_location") == "us-east1"
+        assert "api_key" not in kwargs
+
+    def test_auth_precedence_no_api_key_injection(self, ai_config):
+        """Verifies that api_key is not injected into the auth dict when auth is already set (non-empty)."""
+        from connpy.ai import ai
+        ai_config.config["ai"]["engineer_api_key"] = "legacy-eng-key"
+        ai_config.config["ai"]["architect_api_key"] = "legacy-arch-key"
+        ai_config.config["ai"]["engineer_auth"] = {"vertex_project": "proj-eng"}
+        ai_config.config["ai"]["architect_auth"] = {"vertex_project": "proj-arch"}
+        myai = ai(ai_config)
+        assert myai.engineer_auth == {"vertex_project": "proj-eng"}
+        assert "api_key" not in myai.engineer_auth
+        assert myai.architect_auth == {"vertex_project": "proj-arch"}
+        assert "api_key" not in myai.architect_auth
+
+
 # =========================================================================
 # register_ai_tool tests
 # =========================================================================
@@ -427,12 +498,14 @@ class TestAISessions:
 
     def test_generate_session_id(self, myai):
         session_id = myai._generate_session_id("Any query")
-        # Format: YYYYMMDD-HHMMSS
+        # Format: YYYYMMDD-HHMMSS-suffix
-        assert len(session_id) == 15
+        assert len(session_id) == 20
         assert "-" in session_id
         parts = session_id.split("-")
+        assert len(parts) == 3
         assert len(parts[0]) == 8 # YYYYMMDD
         assert len(parts[1]) == 6 # HHMMSS
+        assert len(parts[2]) == 4 # suffix
 
     def test_save_and_load_session(self, myai):
         history = [

connpy/tests/test_ai_copilot.py

@@ -158,3 +158,245 @@ def test_ingress_task_interception():
         assert called_copilot
         
     asyncio.run(run_test())
+
+def test_build_context_blocks_horizontal_scrolling():
+    from connpy.services.ai_service import AIService
+    svc = AIService(None)
+    
+    node_info = {"prompt": "RP/0/RP0/CPU0:xrd#"}
+    part1 = 'RP/0/RP0/CPU0:xrd#s show interfaces * | inc "rate|is up|escr|test1|test2|test3|test4|test5|teest8|test7|t$'
+    part2 = '|escr|test1|test2|test3|test4|test5|teest8|test7|te                                 s998"show interfaces * | inc "rate|is up|escr|test1|test2|test3|test4|test5|teest8|test7|$'
+    
+    # Test with \r (classic IOS)
+    raw_bytes = (part1 + '\r' + part2).encode()
+    cmd_byte_positions = [(0, None), (len(raw_bytes), None)]
+    
+    blocks = svc.build_context_blocks(raw_bytes, cmd_byte_positions, node_info)
+    assert len(blocks) >= 1
+    start, end, preview = blocks[0]
+    assert "RP/0/RP0/CPU0:xrd# s show interfaces * | inc" in preview
+
+def test_build_context_blocks_horizontal_scrolling_ansi():
+    """Test with CSI cursor repositioning (\\x1B[1G) instead of raw \\r, as used by Cisco IOS XR."""
+    from connpy.services.ai_service import AIService
+    svc = AIService(None)
+    
+    node_info = {"prompt": "RP/0/RP0/CPU0:xrd#"}
+    part1 = 'RP/0/RP0/CPU0:xrd#s show interfaces * | inc "rate|is up|escr|test1|test2|test3|test4|test5|teest8|test7|t'
+    part2 = '$|escr|test1|test2|test3|test4|test5|teest8|test7|te                                 s998"show interfaces * | inc "rate|is up|escr|test1|test2|test3|test4|test5|teest8|test7|$'
+    
+    # Test with \x1B[1G (CSI Cursor Horizontal Absolute - IOS XR)
+    raw_bytes = (part1 + '\x1b[1G' + part2).encode()
+    cmd_byte_positions = [(0, None), (len(raw_bytes), None)]
+    
+    blocks = svc.build_context_blocks(raw_bytes, cmd_byte_positions, node_info)
+    assert len(blocks) >= 1
+    start, end, preview = blocks[0]
+    assert "RP/0/RP0/CPU0:xrd# s show interfaces * | inc" in preview
+
+
+def test_build_context_blocks_cancelled_command():
+    from connpy.services.ai_service import AIService
+    svc = AIService(None)
+    
+    node_info = {"prompt": "router#"}
+    # Command 1: cancelled with Ctrl+C. Command 2: executed successfully.
+    raw_bytes = b"router# show plat\x03\r\nrouter# show ver\r\nrouter# "
+    
+    # 0: initial boundary
+    # 18: Ctrl+C pressed (ends Command 1, marked CANCELLED)
+    # 36: Enter pressed (ends Command 2)
+    cmd_byte_positions = [(0, None), (18, "CANCELLED"), (36, None)]
+    
+    blocks = svc.build_context_blocks(raw_bytes, cmd_byte_positions, node_info)
+    
+    # The cancelled command block (0 to 18) should NOT be registered as a VALID_CMD block.
+    # The block for "show ver" should be registered (starting at 36, ending at current_prompt_pos).
+    # Plus, the final block for "CURRENT CONTEXT".
+    valid_blocks = [b for b in blocks if "CURRENT CONTEXT" not in b[2]]
+    assert len(valid_blocks) == 1
+    assert "show ver" in valid_blocks[0][2]
+    assert "show plat" not in valid_blocks[0][2]
+
+
+def test_copilot_range_mode_filtering():
+    from connpy.cli.terminal_ui import CopilotInterface
+    
+    # We setup dummy raw_bytes with scrolling garbage in the middle:
+    # 0 to 10: "show ip" (VALID_CMD)
+    # 10 to 25: "some scrolling garbage we want to skip"
+    # 25 to 35: "show run" (VALID_CMD)
+    # 35 to 45: "current prompt" (final context block)
+    raw_bytes = b"show ip    garbage_to_skip_here   show run   router#"
+    
+    blocks = [
+        (0, 10, "router# show ip"),
+        (25, 35, "router# show run"),
+        (35, 45, "router#")
+    ]
+    
+    # Mock Config
+    class MockConfig:
+        def __init__(self):
+            self.config = {"ai": {}}
+            self.defaultdir = "/tmp"
+            
+    interface = CopilotInterface(MockConfig())
+    # Ensure default is RANGE mode
+    interface.mode_range = 0
+    interface.mode_single = 1
+    interface.mode_lines = 2
+    
+    captured_buffer = None
+    
+    async def mock_ai_call(active_buffer, question, on_chunk, node_info):
+        nonlocal captured_buffer
+        captured_buffer = active_buffer
+        return {"guide": "Ok", "commands": [], "risk_level": "low"}
+        
+    # Mock PromptSession.prompt_async to ask a question once then exit
+    prompt_calls = 0
+    async def mock_prompt_async(self, *args, **kwargs):
+        nonlocal prompt_calls
+        prompt_calls += 1
+        if prompt_calls == 1:
+            # Simulate pressing Ctrl+Up key twice to expand context range from 1 to 3 commands
+            kb = kwargs.get('key_bindings')
+            if kb:
+                class DummyApp:
+                    def invalidate(self): pass
+                class DummyEvent:
+                    app = DummyApp()
+                
+                # Find and invoke the 'c-up' handler twice
+                for b in kb.bindings:
+                    if any('up' in str(k).lower() for k in b.keys):
+                        b.handler(DummyEvent())
+                        b.handler(DummyEvent())
+            return "how are interfaces looking?"
+        else:
+            raise KeyboardInterrupt
+            
+    with patch('prompt_toolkit.PromptSession.prompt_async', mock_prompt_async):
+        async def run():
+            # Run session
+            return await interface.run_session(
+                raw_bytes=raw_bytes,
+                node_info={"name": "test"},
+                on_ai_call=mock_ai_call,
+                blocks=blocks
+            )
+            
+        asyncio.run(run())
+        
+    # In range mode: it should have concatenated the valid blocks
+    # block[0] is raw_bytes[0:10] => b"show ip    "
+    # block[1] is raw_bytes[25:35] => b"   show run"
+    # block[2] is raw_bytes[35:45] => b"   router#"
+    # Note: raw_bytes[10:25] (garbage) must be excluded!
+    assert captured_buffer is not None
+    assert "garbage_to_skip_here" not in captured_buffer
+    assert "show ip" in captured_buffer
+    assert "show run" in captured_buffer
+
+
+def test_build_context_blocks_pager_scrolling_enter():
+    from connpy.services.ai_service import AIService
+    svc = AIService(None)
+    
+    node_info = {"prompt": "sixwind>"}
+    raw_bytes = (
+        b"sixwind> show configuration | less\r\n"
+        b"line 1 of output\nline 2 of output\n\r"
+        b"line 3 of output\nline 4 of output\n\r"
+        b"line 5 of output\n(END)\x1b[?1049l\x1b[?47l\r\nsixwind> \r\n"
+        b"sixwind> \r\n"
+        b"sixwind> \r\n"
+        b"sixwind> "
+    )
+    cmd_byte_positions = [
+        (0, None),
+        (36, None),
+        (70, None),
+        (105, None),
+        (153, None),
+        (164, None),
+        (175, None),
+        (186, None)
+    ]
+    
+    blocks = svc.build_context_blocks(raw_bytes, cmd_byte_positions, node_info)
+    
+    valid_blocks = [b for b in blocks if "CURRENT CONTEXT" not in b[2]]
+    assert len(valid_blocks) == 1
+    assert "show configuration" in valid_blocks[0][2]
+    assert valid_blocks[0][0] == 36
+    assert valid_blocks[0][1] == 153
+
+
+def test_build_context_blocks_pager_scrolling_space():
+    from connpy.services.ai_service import AIService
+    svc = AIService(None)
+    
+    node_info = {"prompt": "sixwind>"}
+    raw_bytes = (
+        b"sixwind> show configuration | less\r\n"
+        b"line 1 of output\nline 2 of output\n "
+        b"line 3 of output\nline 4 of output\n "
+        b"line 5 of output\n(END)\x1b[?1049l\x1b[?47l\r\n"
+        b"sixwind> \r\n"
+        b"sixwind> \r\n"
+        b"sixwind> \r\n"
+        b"sixwind> "
+    )
+    cmd_byte_positions = [
+        (0, None),
+        (36, None),
+        (144, None),
+        (155, None),
+        (166, None),
+        (177, None)
+    ]
+    
+    blocks = svc.build_context_blocks(raw_bytes, cmd_byte_positions, node_info)
+    
+    valid_blocks = [b for b in blocks if "CURRENT CONTEXT" not in b[2]]
+    assert len(valid_blocks) == 1
+    assert "show configuration" in valid_blocks[0][2]
+    assert valid_blocks[0][0] == 36
+    assert valid_blocks[0][1] == 155
+
+
+def test_build_context_blocks_pager_scrolling_6wind_escapes():
+    from connpy.services.ai_service import AIService
+    svc = AIService(None)
+    
+    node_info = {"prompt": "6WIND-PE1>", "os": "6wind"}
+    raw_bytes = (
+        b"6WIND-PE1> show config running fullpath nodefault\r\n"
+        b"line 1\r\n"
+        b"line 2\r\n"
+        b":\x1b[K\r\x1b[K/ vrf main interface gre gre2 mtu 8400\r\n"
+        b":\x1b[K\x07\r\x1b[K\x1b[?1l\x1b>6WIND-PE1> \r\n"
+        b"6WIND-PE1> \r\n"
+        b"6WIND-PE1> "
+    )
+    cmd_byte_positions = [
+        (0, None),
+        (52, None),
+        (136, None),
+        (177, None),
+        (177, None),
+        (190, None),
+        (203, None)
+    ]
+    
+    blocks = svc.build_context_blocks(raw_bytes, cmd_byte_positions, node_info)
+    
+    valid_blocks = [b for b in blocks if "CURRENT CONTEXT" not in b[2]]
+    assert len(valid_blocks) == 1
+    assert "show config running" in valid_blocks[0][2]
+
+
+
+

connpy/tests/test_completion.py

@@ -65,4 +65,80 @@ class TestGetCwd:
         assert len(dirs_in_result) > 0
 
 
+# =========================================================================
+# Tree completions tests
+# =========================================================================
+
+class TestTreeCompletions:
+    def test_config_auth_completions(self):
+        from connpy.completion import _build_tree, resolve_completion
+        tree = _build_tree([], [], [], {}, "/tmp")
+        # Test config completions
+        config_completions = resolve_completion(["config", ""], tree)
+        assert "--engineer-auth" in config_completions
+        assert "--architect-auth" in config_completions
+
+        # Resolve when --engineer-auth is chosen in config
+        auth_comp = resolve_completion(["config", "--engineer-auth", ""], tree)
+        assert isinstance(auth_comp, list)
+
+        # Loop back check:
+        # e.g., connpy config --engineer-auth some_val
+        # should loop back and resolve to config options
+        loop_back_comp = resolve_completion(["config", "--engineer-auth", "some_val", ""], tree)
+        assert "--architect-auth" in loop_back_comp
+        assert "--engineer-auth" in loop_back_comp
+
+    def test_ai_auth_completions(self):
+        from connpy.completion import _build_tree, resolve_completion
+        tree = _build_tree([], [], [], {}, "/tmp")
+        # Test ai completions
+        ai_completions = resolve_completion(["ai", ""], tree)
+        assert "--engineer-auth" in ai_completions
+        assert "--architect-auth" in ai_completions
+
+        # Resolve after choosing option
+        auth_comp = resolve_completion(["ai", "--engineer-auth", ""], tree)
+        assert isinstance(auth_comp, list)
+
+        # Loop back check:
+        # e.g., connpy ai --engineer-auth some_val
+        # should loop back and resolve to ai options, excluding --engineer-auth
+        loop_back_comp = resolve_completion(["ai", "--engineer-auth", "some_val", ""], tree)
+        assert "--architect-auth" in loop_back_comp
+        assert "--engineer-auth" not in loop_back_comp
+
+    def test_sixwindmcp_plugin_completions(self):
+        from connpy.completion import resolve_completion, get_cwd
+        import importlib.util
+        
+        # Load the testremote/remote_plugins/sixwindmcp.py plugin
+        plugin_path = os.path.join(
+            os.path.dirname(os.path.dirname(os.path.dirname(__file__))),
+            "testremote", "remote_plugins", "sixwindmcp.py"
+        )
+        spec = importlib.util.spec_from_file_location("sixwindmcp", plugin_path)
+        module = importlib.util.module_from_spec(spec)
+        spec.loader.exec_module(module)
+        module.get_cwd = get_cwd
+        
+        plugin_node = module._connpy_tree()
+        assert "--set-path" in plugin_node
+        assert "--path" in plugin_node
+        assert "start" in plugin_node
+        
+        tree = {"sixwindmcp": plugin_node}
+        
+        # Test resolution when --set-path is chosen
+        res = resolve_completion(["sixwindmcp", "--set-path", ""], tree)
+        assert isinstance(res, list)
+        
+        # Loop back check:
+        # e.g., connpy sixwindmcp --set-path /tmp start
+        # should loop back and resolve to plugin options
+        loop_back_comp = resolve_completion(["sixwindmcp", "--set-path", "/tmp", ""], tree)
+        assert "start" in loop_back_comp
+        assert "stop" in loop_back_comp
+
+
 

connpy/tests/test_connapp.py

@@ -246,7 +246,7 @@ def test_plugin_disable(mock_disable, app):
 
 @patch("connpy.services.ai_service.AIService.list_sessions")
 def test_ai_list(mock_list_sessions, app):
-    mock_list_sessions.return_value = [{"id": "1", "title": "t", "created_at": "now", "model": "m"}]
+    mock_list_sessions.return_value = ([{"id": "1", "title": "t", "created_at": "now", "model": "m"}], 1)
     app.start(["ai", "--list"])
     mock_list_sessions.assert_called_once()
 
@@ -262,3 +262,55 @@ def test_type_node_reserved_word(app):
         with pytest.raises(SystemExit) as exc:
             app._type_node("bulk")
         assert exc.value.code == 2
+
+@patch("connpy.services.config_service.ConfigService.update_setting")
+@patch("connpy.services.config_service.ConfigService.get_settings")
+def test_config_auth_inline_json(mock_get_settings, mock_update_setting, app):
+    mock_get_settings.return_value = {"ai": {}}
+    app.start(["config", "--engineer-auth", '{"vertex_project": "test-123"}'])
+    mock_update_setting.assert_called_once()
+    args, kwargs = mock_update_setting.call_args
+    assert args[0] == "ai"
+    assert args[1]["engineer_auth"] == {"vertex_project": "test-123"}
+
+@patch("connpy.services.config_service.ConfigService.update_setting")
+@patch("connpy.services.config_service.ConfigService.get_settings")
+def test_config_auth_inline_yaml(mock_get_settings, mock_update_setting, app):
+    mock_get_settings.return_value = {"ai": {}}
+    app.start(["config", "--architect-auth", 'project: test-yaml'])
+    mock_update_setting.assert_called_once()
+    args, kwargs = mock_update_setting.call_args
+    assert args[0] == "ai"
+    assert args[1]["architect_auth"] == {"project": "test-yaml"}
+
+@patch("connpy.services.config_service.ConfigService.update_setting")
+@patch("connpy.services.config_service.ConfigService.get_settings")
+def test_config_clear_auth(mock_get_settings, mock_update_setting, app):
+    mock_get_settings.return_value = {"ai": {"engineer_auth": {"project": "123"}, "engineer_api_key": "some-key"}}
+    
+    app.start(["config", "--engineer-auth", "clear"])
+    args, kwargs = mock_update_setting.call_args
+    assert "engineer_auth" not in args[1]
+    
+    app.start(["config", "--engineer-api-key", "none"])
+    args, kwargs = mock_update_setting.call_args
+    assert "engineer_api_key" not in args[1]
+
+@patch("os.path.exists")
+@patch("builtins.open")
+@patch("connpy.services.config_service.ConfigService.update_setting")
+@patch("connpy.services.config_service.ConfigService.get_settings")
+def test_config_auth_file_path(mock_get_settings, mock_update_setting, mock_open, mock_exists, app):
+    mock_get_settings.return_value = {"ai": {}}
+    mock_exists.side_effect = lambda p: True if p == "/path/to/creds.json" else False
+    mock_file = MagicMock()
+    mock_file.read.return_value = '{"vertex_project": "file-project"}'
+    mock_open.return_value.__enter__.return_value = mock_file
+    
+    app.start(["config", "--engineer-auth", "/path/to/creds.json"])
+    mock_update_setting.assert_called_once()
+    args, kwargs = mock_update_setting.call_args
+    assert args[0] == "ai"
+    assert args[1]["engineer_auth"] == {"vertex_project": "file-project"}
+
+

connpy/tests/test_grpc_auth.py

@@ -0,0 +1,136 @@
+"""
+Tests for gRPC auth serialization/deserialization (engineer_auth, architect_auth, provider auth).
+
+These tests verify that:
+1. to_struct/from_struct round-trips correctly for auth dicts.
+2. AIStub.ask() correctly serializes engineer_auth and architect_auth into AskRequest.
+3. AIServicer.ask() correctly deserializes them and passes them to the service.
+4. AIStub.configure_provider() serializes auth into ProviderRequest.
+5. AIServicer.configure_provider() deserializes auth and forwards it to the service.
+"""
+import pytest
+from unittest.mock import MagicMock, patch, call
+from connpy.grpc_layer import connpy_pb2
+from connpy.grpc_layer.utils import to_struct, from_struct
+
+
+# --- Unit: Struct round-trip ---
+
+class TestStructRoundTrip:
+    def test_simple_dict(self):
+        d = {"api_key": "secret", "region": "us-east-1"}
+        assert from_struct(to_struct(d)) == d
+
+    def test_nested_dict(self):
+        d = {"vertex_project": "my-project", "vertex_location": "us-central1", "nested": {"key": "val"}}
+        assert from_struct(to_struct(d)) == d
+
+    def test_empty_dict(self):
+        assert from_struct(to_struct({})) == {}
+
+    def test_none_returns_empty(self):
+        assert from_struct(to_struct(None)) == {}
+
+
+# --- Unit: AskRequest Struct fields ---
+
+class TestAskRequestStructFields:
+    def test_engineer_auth_round_trip(self):
+        auth = {"vertex_project": "proj", "vertex_location": "us-central1"}
+        req = connpy_pb2.AskRequest(input_text="hi")
+        req.engineer_auth.CopyFrom(to_struct(auth))
+        assert from_struct(req.engineer_auth) == auth
+
+    def test_architect_auth_round_trip(self):
+        auth = {"api_key": "sk-abc", "base_url": "https://custom.api/v1"}
+        req = connpy_pb2.AskRequest(input_text="hi")
+        req.architect_auth.CopyFrom(to_struct(auth))
+        assert from_struct(req.architect_auth) == auth
+
+    def test_has_field_false_when_unset(self):
+        req = connpy_pb2.AskRequest(input_text="hi")
+        assert not req.HasField("engineer_auth")
+        assert not req.HasField("architect_auth")
+
+    def test_has_field_true_when_set(self):
+        req = connpy_pb2.AskRequest(input_text="hi")
+        req.engineer_auth.CopyFrom(to_struct({"k": "v"}))
+        assert req.HasField("engineer_auth")
+
+
+# --- Unit: ProviderRequest Struct field ---
+
+class TestProviderRequestStructField:
+    def test_auth_round_trip(self):
+        auth = {"vertex_project": "proj", "vertex_location": "eu-west1"}
+        req = connpy_pb2.ProviderRequest(provider="vertex", model="gemini-pro")
+        req.auth.CopyFrom(to_struct(auth))
+        assert from_struct(req.auth) == auth
+
+    def test_has_field_false_when_unset(self):
+        req = connpy_pb2.ProviderRequest(provider="openai", model="gpt-4o")
+        assert not req.HasField("auth")
+
+    def test_has_field_true_when_set(self):
+        req = connpy_pb2.ProviderRequest(provider="vertex")
+        req.auth.CopyFrom(to_struct({"vertex_project": "p"}))
+        assert req.HasField("auth")
+
+
+# --- Integration: Server deserializes auth and passes to service ---
+
+class TestAIServicerAuthDeserialization:
+    @pytest.fixture
+    def servicer(self, populated_config):
+        from connpy.grpc_layer.server import AIServicer
+        return AIServicer(populated_config)
+
+    def test_configure_provider_passes_auth_to_service(self, servicer):
+        auth = {"vertex_project": "my-proj", "vertex_location": "us-central1"}
+        req = connpy_pb2.ProviderRequest(provider="vertex", model="gemini/gemini-pro", api_key="")
+        req.auth.CopyFrom(to_struct(auth))
+
+        with patch.object(servicer.service, "configure_provider") as mock_cp:
+            mock_context = MagicMock()
+            servicer.configure_provider(req, mock_context)
+            mock_cp.assert_called_once_with("vertex", "gemini/gemini-pro", "", auth=auth)
+
+    def test_configure_provider_no_auth(self, servicer):
+        req = connpy_pb2.ProviderRequest(provider="openai", model="gpt-4o", api_key="sk-test")
+
+        with patch.object(servicer.service, "configure_provider") as mock_cp:
+            mock_context = MagicMock()
+            servicer.configure_provider(req, mock_context)
+            mock_cp.assert_called_once_with("openai", "gpt-4o", "sk-test", auth=None)
+
+
+# --- Integration: Stub serializes auth into request ---
+
+class TestAIStubAuthSerialization:
+    @pytest.fixture
+    def ai_stub(self):
+        from connpy.grpc_layer.stubs import AIStub
+        mock_channel = MagicMock()
+        stub = AIStub(mock_channel, "localhost:8048")
+        return stub
+
+    def test_configure_provider_with_auth_serializes_struct(self, ai_stub):
+        auth = {"vertex_project": "proj", "vertex_location": "us-central1"}
+        ai_stub.stub.configure_provider = MagicMock()
+
+        ai_stub.configure_provider("vertex", model="gemini/gemini-pro", auth=auth)
+
+        ai_stub.stub.configure_provider.assert_called_once()
+        sent_req = ai_stub.stub.configure_provider.call_args[0][0]
+        assert sent_req.provider == "vertex"
+        assert sent_req.model == "gemini/gemini-pro"
+        assert sent_req.HasField("auth")
+        assert from_struct(sent_req.auth) == auth
+
+    def test_configure_provider_without_auth_no_struct(self, ai_stub):
+        ai_stub.stub.configure_provider = MagicMock()
+
+        ai_stub.configure_provider("openai", model="gpt-4o", api_key="sk-x")
+
+        sent_req = ai_stub.stub.configure_provider.call_args[0][0]
+        assert not sent_req.HasField("auth")

connpy/tests/test_grpc_layer.py

@@ -120,6 +120,7 @@ class TestGRPCIntegration:
         connpy_pb2_grpc.add_ConfigServiceServicer_to_server(server.ConfigServicer(populated_config), srv)
         connpy_pb2_grpc.add_ExecutionServiceServicer_to_server(server.ExecutionServicer(populated_config), srv)
         connpy_pb2_grpc.add_ImportExportServiceServicer_to_server(server.ImportExportServicer(populated_config), srv)
+        connpy_pb2_grpc.add_AIServiceServicer_to_server(server.AIServicer(populated_config), srv)
         
         port = srv.add_insecure_port('127.0.0.1:0')
         srv.start()
@@ -143,6 +144,10 @@ class TestGRPCIntegration:
     def config_stub(self, channel):
         return stubs.ConfigStub(channel, "localhost")
 
+    @pytest.fixture
+    def ai_stub(self, channel):
+        return stubs.AIStub(channel, "localhost")
+
     def test_list_nodes_integration(self, node_stub):
         nodes = node_stub.list_nodes()
         assert "router1" in nodes
@@ -170,6 +175,12 @@ class TestGRPCIntegration:
         settings = config_stub.get_settings()
         assert settings["idletime"] == 99
 
+    def test_list_mcp_servers_integration(self, ai_stub):
+        ai_stub.configure_mcp("test-mcp", url="http://localhost:8080", enabled=True)
+        servers = ai_stub.list_mcp_servers()
+        assert "test-mcp" in servers
+        assert servers["test-mcp"]["url"] == "http://localhost:8080"
+
     def test_add_delete_node_integration(self, node_stub):
         node_stub.add_node("integration-test-node", {"host": "9.9.9.9"})
         assert "integration-test-node" in node_stub.list_nodes()

connpy/tests/test_utils.py

@@ -0,0 +1,32 @@
+import pytest
+from connpy.utils import log_cleaner
+
+def test_log_cleaner_empty():
+    assert log_cleaner("") == ""
+    assert log_cleaner(None) == ""
+
+def test_log_cleaner_plain_text():
+    assert log_cleaner("hello world") == "hello world"
+
+def test_log_cleaner_ansi_colors():
+    # \x1b[31m is red, \x1b[0m is reset
+    assert log_cleaner("\x1b[31mhello\x1b[0m world") == "hello world"
+
+def test_log_cleaner_osc_window_title():
+    # Set window title OSC: \x1b]0;my title\x07 followed by prompt
+    sample = "\x1b]0;fluzzi32@norman: ~\x07fluzzi32@norman:~$"
+    assert log_cleaner(sample) == "fluzzi32@norman:~$"
+
+def test_log_cleaner_osc_with_st_terminator():
+    # OSC can also be terminated by \x1b\\ (ST)
+    sample = "\x1b]0;some title\x1b\\my_prompt>"
+    assert log_cleaner(sample) == "my_prompt>"
+
+def test_log_cleaner_mixed_ansi_and_osc():
+    sample = "\x1b]0;title\x07\x1b[32muser@host\x1b[0m:\x1b[34m/path\x1b[0m$ "
+    assert log_cleaner(sample) == "user@host:/path$"
+
+def test_log_cleaner_carriage_return_and_backspace():
+    # Test that standard control sequences like \r and \b still work as expected
+    assert log_cleaner("hello\rworld") == "world"
+    assert log_cleaner("hell\bo") == "helo"

connpy/utils.py

@@ -7,11 +7,14 @@ def log_cleaner(data: str) -> str:
     if not data:
         return ""
             
+    # Remove OSC (Operating System Command) sequences (e.g., set window title \x1b]0;...\x07)
+    data = re.sub(r'\x1b\][^\x07\x1b]*(?:\x07|\x1b\\)', '', data)
+
     lines = data.split('\n')
     cleaned_lines = []
     
     # Regex to capture: ANSI sequences, control characters (\r, \b, etc), and plain text chunks
-    token_re = re.compile(r'(\x1B(?:[@-Z\\-_]|\[[0-?]*[ -/ ]*[@-~])|\r|\b|\x7f|[\x00-\x1F]|[^\x1B\r\b\x7f\x00-\x1F]+)')
+    token_re = re.compile(r'(\x1B(?:[\x30-\x5A\x5C-\x7E]|\[[0-?]*[ -/ ]*[@-~])|\r|\b|\x7f|[\x00-\x1F]|[^\x1B\r\b\x7f\x00-\x1F]+)')
     
     for line in lines:
         buffer = []

docs/connpy/cli/ai_handler.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.ai_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -61,13 +61,22 @@ el.replaceWith(d);
 
     def dispatch(self, args):
         if args.list_sessions:
-            sessions = self.app.services.ai.list_sessions()
+            limit = 20 if not getattr(args, &#34;all&#34;, False) else None
+            sessions, total = self.app.services.ai.list_sessions(limit=limit)
             if not sessions:
                 printer.info(&#34;No saved AI sessions found.&#34;)
                 return
+            
             columns = [&#34;ID&#34;, &#34;Title&#34;, &#34;Created At&#34;, &#34;Model&#34;]
             rows = [[s[&#34;id&#34;], s[&#34;title&#34;], s[&#34;created_at&#34;], s[&#34;model&#34;]] for s in sessions]
-            printer.table(&#34;AI Persisted Sessions&#34;, columns, rows)
+            
+            title = &#34;AI Persisted Sessions&#34;
+            if limit and total &gt; limit:
+                title += f&#34; (Showing last {limit} of {total})&#34;
+                
+            printer.table(title, columns, rows)
+            if limit and total &gt; limit:
+                printer.info(f&#34;Use &#39;--list --all&#39; to see all {total} sessions.&#34;)
             return
             
         if args.delete_session:
@@ -84,7 +93,7 @@ el.replaceWith(d);
         # Determinar session_id para retomar
         session_id = None
         if args.resume:
-            sessions = self.app.services.ai.list_sessions()
+            sessions, _ = self.app.services.ai.list_sessions()
             session_id = sessions[0][&#34;id&#34;] if sessions else None
             if not session_id:
                 printer.warning(&#34;No previous session found to resume.&#34;)
@@ -103,15 +112,22 @@ el.replaceWith(d);
             elif settings.get(key):
                 arguments[key] = settings.get(key)
 
+        for key in [&#34;engineer_auth&#34;, &#34;architect_auth&#34;]:
+            cli_val = getattr(args, key, None)
+            if cli_val:
+                arguments[key] = self._parse_auth_value(cli_val[0])
+            elif settings.get(key):
+                arguments[key] = settings.get(key)
+        
         # Check keys only if running in local mode (not remote)
         if getattr(self.app.services, &#34;mode&#34;, &#34;local&#34;) == &#34;local&#34;:
-            if not arguments.get(&#34;engineer_api_key&#34;):
+            if not arguments.get(&#34;engineer_api_key&#34;) and not arguments.get(&#34;engineer_auth&#34;):
-                printer.error(&#34;Engineer API key not configured. The chat cannot start.&#34;)
+                printer.error(&#34;Engineer API key/auth not configured. The chat cannot start.&#34;)
-                printer.info(&#34;Use &#39;connpy config --engineer-api-key &lt;key&gt;&#39; to set it.&#34;)
+                printer.info(&#34;Use &#39;connpy config --engineer-api-key &lt;key&gt;&#39; or &#39;connpy config --engineer-auth &lt;auth&gt;&#39; to set it.&#34;)
                 sys.exit(1)
-            if not arguments.get(&#34;architect_api_key&#34;):
+            if not arguments.get(&#34;architect_api_key&#34;) and not arguments.get(&#34;architect_auth&#34;):
-                printer.warning(&#34;Architect API key not configured. Architect will be unavailable.&#34;)
+                printer.warning(&#34;Architect API key/auth not configured. Architect will be unavailable.&#34;)
-                printer.info(&#34;Use &#39;connpy config --architect-api-key &lt;key&gt;&#39; to enable it.&#34;)
+                printer.info(&#34;Use &#39;connpy config --architect-api-key &lt;key&gt;&#39; or &#39;connpy config --architect-auth &lt;auth&gt;&#39; to enable it.&#34;)
 
         # El resto de la interacción el CLI la maneja con el agente subyacente
         self.app.myai = self.app.services.ai
@@ -148,7 +164,7 @@ el.replaceWith(d);
                 if history:
                     mdprint(f&#34;[debug]Analyzing {len(history)} previous messages...[/debug]\n&#34;)
             else:
-                printer.error(f&#34;Could not load session {session_id}. Starting clean.&#34;)
+                printer.info(f&#34;Session &#39;{session_id}&#39; not found. Starting clean.&#34;)
         
         if not history:
             mdprint(Rule(style=&#34;engineer&#34;))
@@ -162,7 +178,7 @@ el.replaceWith(d);
                 if user_query.lower() in [&#39;exit&#39;, &#39;quit&#39;, &#39;bye&#39;, &#39;cancel&#39;]: break
                 
                 with console.status(&#34;[ai_status]Agent is thinking...&#34;) as status:
-                    result = self.app.myai.ask(user_query, chat_history=history, status=status, debug=args.debug, trust=args.trust, **self.ai_overrides)
+                    result = self.app.myai.ask(user_query, chat_history=history, status=status, debug=args.debug, trust=args.trust, session_id=session_id, **self.ai_overrides)
                 
                 new_history = result.get(&#34;chat_history&#34;)
                 if new_history is not None:
@@ -193,8 +209,7 @@ el.replaceWith(d);
             action = mcp_args[0].lower()
             
             if action == &#34;list&#34;:
-                settings = self.app.services.config_svc.get_settings()
+                mcp_servers = self.app.services.ai.list_mcp_servers()
-                mcp_servers = settings.get(&#34;ai&#34;, {}).get(&#34;mcp_servers&#34;, {})
                 if not mcp_servers:
                     printer.info(&#34;No MCP servers configured.&#34;)
                 else:
@@ -259,8 +274,7 @@ el.replaceWith(d);
             from .forms import Forms
             self.app.cli_forms = Forms(self.app)
             
-        settings = self.app.services.config_svc.get_settings()
+        mcp_servers = self.app.services.ai.list_mcp_servers()
-        mcp_servers = settings.get(&#34;ai&#34;, {}).get(&#34;mcp_servers&#34;, {})
         
         result = self.app.cli_forms.mcp_wizard(mcp_servers)
         if not result:
@@ -294,7 +308,37 @@ el.replaceWith(d);
                 printer.success(f&#34;MCP server &#39;{result[&#39;name&#39;]}&#39; removed.&#34;)
                 
         except Exception as e:
-            printer.error(str(e))</code></pre>
+            printer.error(str(e))
+
+    def _parse_auth_value(self, value):
+        if not value or value.lower() in [&#34;none&#34;, &#34;clear&#34;]:
+            return None
+        import os
+        import yaml
+        import json
+        if os.path.exists(value):
+            try:
+                with open(value, &#34;r&#34;) as f:
+                    content = f.read()
+                try:
+                    return json.loads(content)
+                except ValueError:
+                    return yaml.safe_load(content)
+            except Exception as e:
+                printer.error(f&#34;Failed to read/parse auth file &#39;{value}&#39;: {e}&#34;)
+                sys.exit(1)
+        
+        try:
+            return json.loads(value)
+        except ValueError:
+            try:
+                parsed = yaml.safe_load(value)
+                if isinstance(parsed, dict):
+                    return parsed
+                raise ValueError()
+            except Exception:
+                printer.error(&#34;Auth parameter must be a valid JSON/YAML string, or a path to a JSON/YAML file.&#34;)
+                sys.exit(1)</code></pre>
 </details>
 <div class="desc"></div>
 <h3>Methods</h3>
@@ -316,8 +360,7 @@ el.replaceWith(d);
         action = mcp_args[0].lower()
         
         if action == &#34;list&#34;:
-            settings = self.app.services.config_svc.get_settings()
+            mcp_servers = self.app.services.ai.list_mcp_servers()
-            mcp_servers = settings.get(&#34;ai&#34;, {}).get(&#34;mcp_servers&#34;, {})
             if not mcp_servers:
                 printer.info(&#34;No MCP servers configured.&#34;)
             else:
@@ -382,8 +425,7 @@ el.replaceWith(d);
         from .forms import Forms
         self.app.cli_forms = Forms(self.app)
         
-    settings = self.app.services.config_svc.get_settings()
+    mcp_servers = self.app.services.ai.list_mcp_servers()
-    mcp_servers = settings.get(&#34;ai&#34;, {}).get(&#34;mcp_servers&#34;, {})
     
     result = self.app.cli_forms.mcp_wizard(mcp_servers)
     if not result:
@@ -431,13 +473,22 @@ el.replaceWith(d);
 </summary>
 <pre><code class="python">def dispatch(self, args):
     if args.list_sessions:
-        sessions = self.app.services.ai.list_sessions()
+        limit = 20 if not getattr(args, &#34;all&#34;, False) else None
+        sessions, total = self.app.services.ai.list_sessions(limit=limit)
         if not sessions:
             printer.info(&#34;No saved AI sessions found.&#34;)
             return
+        
         columns = [&#34;ID&#34;, &#34;Title&#34;, &#34;Created At&#34;, &#34;Model&#34;]
         rows = [[s[&#34;id&#34;], s[&#34;title&#34;], s[&#34;created_at&#34;], s[&#34;model&#34;]] for s in sessions]
-        printer.table(&#34;AI Persisted Sessions&#34;, columns, rows)
+        
+        title = &#34;AI Persisted Sessions&#34;
+        if limit and total &gt; limit:
+            title += f&#34; (Showing last {limit} of {total})&#34;
+            
+        printer.table(title, columns, rows)
+        if limit and total &gt; limit:
+            printer.info(f&#34;Use &#39;--list --all&#39; to see all {total} sessions.&#34;)
         return
         
     if args.delete_session:
@@ -454,7 +505,7 @@ el.replaceWith(d);
     # Determinar session_id para retomar
     session_id = None
     if args.resume:
-        sessions = self.app.services.ai.list_sessions()
+        sessions, _ = self.app.services.ai.list_sessions()
         session_id = sessions[0][&#34;id&#34;] if sessions else None
         if not session_id:
             printer.warning(&#34;No previous session found to resume.&#34;)
@@ -473,15 +524,22 @@ el.replaceWith(d);
         elif settings.get(key):
             arguments[key] = settings.get(key)
 
+    for key in [&#34;engineer_auth&#34;, &#34;architect_auth&#34;]:
+        cli_val = getattr(args, key, None)
+        if cli_val:
+            arguments[key] = self._parse_auth_value(cli_val[0])
+        elif settings.get(key):
+            arguments[key] = settings.get(key)
+    
     # Check keys only if running in local mode (not remote)
     if getattr(self.app.services, &#34;mode&#34;, &#34;local&#34;) == &#34;local&#34;:
-        if not arguments.get(&#34;engineer_api_key&#34;):
+        if not arguments.get(&#34;engineer_api_key&#34;) and not arguments.get(&#34;engineer_auth&#34;):
-            printer.error(&#34;Engineer API key not configured. The chat cannot start.&#34;)
+            printer.error(&#34;Engineer API key/auth not configured. The chat cannot start.&#34;)
-            printer.info(&#34;Use &#39;connpy config --engineer-api-key &lt;key&gt;&#39; to set it.&#34;)
+            printer.info(&#34;Use &#39;connpy config --engineer-api-key &lt;key&gt;&#39; or &#39;connpy config --engineer-auth &lt;auth&gt;&#39; to set it.&#34;)
             sys.exit(1)
-        if not arguments.get(&#34;architect_api_key&#34;):
+        if not arguments.get(&#34;architect_api_key&#34;) and not arguments.get(&#34;architect_auth&#34;):
-            printer.warning(&#34;Architect API key not configured. Architect will be unavailable.&#34;)
+            printer.warning(&#34;Architect API key/auth not configured. Architect will be unavailable.&#34;)
-            printer.info(&#34;Use &#39;connpy config --architect-api-key &lt;key&gt;&#39; to enable it.&#34;)
+            printer.info(&#34;Use &#39;connpy config --architect-api-key &lt;key&gt;&#39; or &#39;connpy config --architect-auth &lt;auth&gt;&#39; to enable it.&#34;)
 
     # El resto de la interacción el CLI la maneja con el agente subyacente
     self.app.myai = self.app.services.ai
@@ -512,7 +570,7 @@ el.replaceWith(d);
             if history:
                 mdprint(f&#34;[debug]Analyzing {len(history)} previous messages...[/debug]\n&#34;)
         else:
-            printer.error(f&#34;Could not load session {session_id}. Starting clean.&#34;)
+            printer.info(f&#34;Session &#39;{session_id}&#39; not found. Starting clean.&#34;)
     
     if not history:
         mdprint(Rule(style=&#34;engineer&#34;))
@@ -526,7 +584,7 @@ el.replaceWith(d);
             if user_query.lower() in [&#39;exit&#39;, &#39;quit&#39;, &#39;bye&#39;, &#39;cancel&#39;]: break
             
             with console.status(&#34;[ai_status]Agent is thinking...&#34;) as status:
-                result = self.app.myai.ask(user_query, chat_history=history, status=status, debug=args.debug, trust=args.trust, **self.ai_overrides)
+                result = self.app.myai.ask(user_query, chat_history=history, status=status, debug=args.debug, trust=args.trust, session_id=session_id, **self.ai_overrides)
             
             new_history = result.get(&#34;chat_history&#34;)
             if new_history is not None:
@@ -608,7 +666,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/api_handler.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.api_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -193,7 +193,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/config_handler.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.config_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -70,8 +70,10 @@ el.replaceWith(d);
             &#34;theme&#34;: self.set_theme,
             &#34;engineer_model&#34;: self.set_ai_config,
             &#34;engineer_api_key&#34;: self.set_ai_config,
+            &#34;engineer_auth&#34;: self.set_ai_config,
             &#34;architect_model&#34;: self.set_ai_config,
             &#34;architect_api_key&#34;: self.set_ai_config,
+            &#34;architect_auth&#34;: self.set_ai_config,
             &#34;trusted_commands&#34;: self.set_ai_config,
             &#34;service_mode&#34;: self.set_service_mode,
             &#34;remote_host&#34;: self.set_remote_host,
@@ -178,11 +180,59 @@ el.replaceWith(d);
         try:
             settings = self.app.services.config_svc.get_settings()
             aiconfig = settings.get(&#34;ai&#34;, {})
-            aiconfig[args.command] = args.data[0]
+            val = args.data[0]
+            
+            # Check for unset/clear request
+            if val.lower() in [&#34;none&#34;, &#34;clear&#34;, &#34;&#34;]:
+                if args.command in aiconfig:
+                    del aiconfig[args.command]
+            else:
+                # If configuring auth, parse as dictionary (JSON/YAML or file path)
+                if args.command in [&#34;engineer_auth&#34;, &#34;architect_auth&#34;]:
+                    parsed_val = self._parse_auth_value(val)
+                    if parsed_val is not None:
+                        aiconfig[args.command] = parsed_val
+                    else:
+                        if args.command in aiconfig:
+                            del aiconfig[args.command]
+                else:
+                    aiconfig[args.command] = val
+            
             self.app.services.config_svc.update_setting(&#34;ai&#34;, aiconfig)
             printer.success(&#34;Config saved&#34;)
-        except ConnpyError as e:
+        except (ConnpyError, InvalidConfigurationError) as e:
-            printer.error(str(e))</code></pre>
+            printer.error(str(e))
+
+    def _parse_auth_value(self, value):
+        if value.lower() in [&#34;none&#34;, &#34;clear&#34;, &#34;&#34;]:
+            return None
+        
+        # Check if it&#39;s a file path
+        import os
+        if os.path.exists(value):
+            try:
+                with open(value, &#34;r&#34;) as f:
+                    content = f.read()
+                import json
+                try:
+                    return json.loads(content)
+                except ValueError:
+                    return yaml.safe_load(content)
+            except Exception as e:
+                raise InvalidConfigurationError(f&#34;Failed to read/parse auth file &#39;{value}&#39;: {e}&#34;)
+        
+        # Try parsing as inline JSON/YAML
+        try:
+            import json
+            return json.loads(value)
+        except ValueError:
+            try:
+                parsed = yaml.safe_load(value)
+                if isinstance(parsed, dict):
+                    return parsed
+                raise ValueError()
+            except Exception:
+                raise InvalidConfigurationError(&#34;Auth parameter must be a valid JSON/YAML string, or a path to a JSON/YAML file.&#34;)</code></pre>
 </details>
 <div class="desc"></div>
 <h3>Methods</h3>
@@ -206,8 +256,10 @@ el.replaceWith(d);
         &#34;theme&#34;: self.set_theme,
         &#34;engineer_model&#34;: self.set_ai_config,
         &#34;engineer_api_key&#34;: self.set_ai_config,
+        &#34;engineer_auth&#34;: self.set_ai_config,
         &#34;architect_model&#34;: self.set_ai_config,
         &#34;architect_api_key&#34;: self.set_ai_config,
+        &#34;architect_auth&#34;: self.set_ai_config,
         &#34;trusted_commands&#34;: self.set_ai_config,
         &#34;service_mode&#34;: self.set_service_mode,
         &#34;remote_host&#34;: self.set_remote_host,
@@ -234,10 +286,27 @@ el.replaceWith(d);
     try:
         settings = self.app.services.config_svc.get_settings()
         aiconfig = settings.get(&#34;ai&#34;, {})
-        aiconfig[args.command] = args.data[0]
+        val = args.data[0]
+        
+        # Check for unset/clear request
+        if val.lower() in [&#34;none&#34;, &#34;clear&#34;, &#34;&#34;]:
+            if args.command in aiconfig:
+                del aiconfig[args.command]
+        else:
+            # If configuring auth, parse as dictionary (JSON/YAML or file path)
+            if args.command in [&#34;engineer_auth&#34;, &#34;architect_auth&#34;]:
+                parsed_val = self._parse_auth_value(val)
+                if parsed_val is not None:
+                    aiconfig[args.command] = parsed_val
+                else:
+                    if args.command in aiconfig:
+                        del aiconfig[args.command]
+            else:
+                aiconfig[args.command] = val
+        
         self.app.services.config_svc.update_setting(&#34;ai&#34;, aiconfig)
         printer.success(&#34;Config saved&#34;)
-    except ConnpyError as e:
+    except (ConnpyError, InvalidConfigurationError) as e:
         printer.error(str(e))</code></pre>
 </details>
 <div class="desc"></div>
@@ -482,7 +551,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/context_handler.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.context_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -249,7 +249,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/forms.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.forms API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -690,7 +690,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/help_text.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.help_text API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -303,7 +303,7 @@ tasks:
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/helpers.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.helpers API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -69,7 +69,7 @@ el.replaceWith(d);
             return answer[0]
     else:
         questions = [inquirer.List(name, message=&#34;Pick {} to {}:&#34;.format(name,action), choices=list_, carousel=True)]
-        answer = inquirer.prompt(questions)
+        answer = inquirer.prompt(questions, theme=theme)
         if answer == None:
             return None
         else:
@@ -115,6 +115,65 @@ el.replaceWith(d);
 </details>
 <div class="desc"></div>
 </dd>
+<dt id="connpy.cli.helpers.get_theme"><code class="name flex">
+<span>def <span class="ident">get_theme</span></span>(<span>)</span>
+</code></dt>
+<dd>
+<details class="source">
+<summary>
+<span>Expand source code</span>
+</summary>
+<pre><code class="python">def get_theme():
+    &#34;&#34;&#34;Returns a fresh instance of the theme with current colors.&#34;&#34;&#34;
+    return ConnpyTheme()</code></pre>
+</details>
+<div class="desc"><p>Returns a fresh instance of the theme with current colors.</p></div>
+</dd>
+<dt id="connpy.cli.helpers.hex_to_blessed"><code class="name flex">
+<span>def <span class="ident">hex_to_blessed</span></span>(<span>hex_str)</span>
+</code></dt>
+<dd>
+<details class="source">
+<summary>
+<span>Expand source code</span>
+</summary>
+<pre><code class="python">def hex_to_blessed(hex_str):
+    &#34;&#34;&#34;Convert hex color string to blessed/ansi format.&#34;&#34;&#34;
+    if not hex_str or not isinstance(hex_str, str):
+        return term.normal
+    
+    # Check for bold prefix
+    prefix = &#34;&#34;
+    if hex_str.startswith(&#39;bold &#39;):
+        prefix = term.bold
+        hex_str = hex_str.replace(&#39;bold &#39;, &#39;&#39;).strip()
+    
+    # If it&#39;s a standard color name
+    if not hex_str.startswith(&#39;#&#39;):
+        return prefix + getattr(term, hex_str, term.normal)
+    
+    # Parse hex
+    try:
+        h = hex_str.lstrip(&#39;#&#39;)
+        if len(h) == 3:
+            h = &#39;&#39;.join([c*2 for c in h])
+        r = int(h[0:2], 16)
+        g = int(h[2:4], 16)
+        b = int(h[4:6], 16)
+        
+        # Try RGB, fallback to standard cyan if it fails or returns empty
+        try:
+            c = term.color_rgb(r, g, b)
+            if not c: # Some terms return empty for RGB
+                return prefix + term.cyan
+            return prefix + c
+        except:
+            return prefix + term.cyan
+    except:
+        return prefix + term.normal</code></pre>
+</details>
+<div class="desc"><p>Convert hex color string to blessed/ansi format.</p></div>
+</dd>
 <dt id="connpy.cli.helpers.nodes_completer"><code class="name flex">
 <span>def <span class="ident">nodes_completer</span></span>(<span>prefix, parsed_args, **kwargs)</span>
 </code></dt>
@@ -181,6 +240,61 @@ el.replaceWith(d);
 </dl>
 </section>
 <section>
+<h2 class="section-title" id="header-classes">Classes</h2>
+<dl>
+<dt id="connpy.cli.helpers.ConnpyTheme"><code class="flex name class">
+<span>class <span class="ident">ConnpyTheme</span></span>
+</code></dt>
+<dd>
+<details class="source">
+<summary>
+<span>Expand source code</span>
+</summary>
+<pre><code class="python">class ConnpyTheme(Default):
+    def __init__(self):
+        super().__init__()
+        try:
+            from ..printer import _global_active_styles
+            # Use user_prompt as primary accent, fallback to info/cyan
+            accent = _global_active_styles.get(&#34;user_prompt&#34;, _global_active_styles.get(&#34;info&#34;, &#34;cyan&#34;))
+            accent_color = hex_to_blessed(accent)
+            
+            self.Question.mark_color = accent_color
+            self.List.selection_color = accent_color
+            self.List.selection_cursor = &#34;&gt;&#34;
+        except:
+            # Absolute fallback to standard cyan
+            self.Question.mark_color = term.cyan
+            self.List.selection_color = term.bold_cyan
+            self.List.selection_cursor = &#34;&gt;&#34;</code></pre>
+</details>
+<div class="desc"></div>
+<h3>Ancestors</h3>
+<ul class="hlist">
+<li>inquirer.themes.Default</li>
+<li>inquirer.themes.Theme</li>
+</ul>
+</dd>
+<dt id="connpy.cli.helpers.ThemeProxy"><code class="flex name class">
+<span>class <span class="ident">ThemeProxy</span></span>
+</code></dt>
+<dd>
+<details class="source">
+<summary>
+<span>Expand source code</span>
+</summary>
+<pre><code class="python">class ThemeProxy:
+    &#34;&#34;&#34;Proxy to ensure theme colors are resolved at runtime.&#34;&#34;&#34;
+    def __getattr__(self, name):
+        return getattr(get_theme(), name)
+    def __iter__(self):
+        return iter(get_theme())
+    def __getitem__(self, item):
+        return get_theme()[item]</code></pre>
+</details>
+<div class="desc"><p>Proxy to ensure theme colors are resolved at runtime.</p></div>
+</dd>
+</dl>
 </section>
 </article>
 <nav id="sidebar">
@@ -198,16 +312,28 @@ el.replaceWith(d);
 <li><code><a title="connpy.cli.helpers.choose" href="#connpy.cli.helpers.choose">choose</a></code></li>
 <li><code><a title="connpy.cli.helpers.folders_completer" href="#connpy.cli.helpers.folders_completer">folders_completer</a></code></li>
 <li><code><a title="connpy.cli.helpers.get_config_dir" href="#connpy.cli.helpers.get_config_dir">get_config_dir</a></code></li>
+<li><code><a title="connpy.cli.helpers.get_theme" href="#connpy.cli.helpers.get_theme">get_theme</a></code></li>
+<li><code><a title="connpy.cli.helpers.hex_to_blessed" href="#connpy.cli.helpers.hex_to_blessed">hex_to_blessed</a></code></li>
 <li><code><a title="connpy.cli.helpers.nodes_completer" href="#connpy.cli.helpers.nodes_completer">nodes_completer</a></code></li>
 <li><code><a title="connpy.cli.helpers.profiles_completer" href="#connpy.cli.helpers.profiles_completer">profiles_completer</a></code></li>
 <li><code><a title="connpy.cli.helpers.toplevel_completer" href="#connpy.cli.helpers.toplevel_completer">toplevel_completer</a></code></li>
 </ul>
 </li>
+<li><h3><a href="#header-classes">Classes</a></h3>
+<ul>
+<li>
+<h4><code><a title="connpy.cli.helpers.ConnpyTheme" href="#connpy.cli.helpers.ConnpyTheme">ConnpyTheme</a></code></h4>
+</li>
+<li>
+<h4><code><a title="connpy.cli.helpers.ThemeProxy" href="#connpy.cli.helpers.ThemeProxy">ThemeProxy</a></code></h4>
+</li>
+</ul>
+</li>
 </ul>
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/import_export_handler.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.import_export_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -272,7 +272,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/index.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -142,7 +142,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/node_handler.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.node_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -606,7 +606,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/plugin_handler.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.plugin_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -385,7 +385,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/profile_handler.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.profile_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -314,7 +314,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/run_handler.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.run_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -454,7 +454,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/sync_handler.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.sync_handler API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -427,7 +427,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/terminal_ui.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.terminal_ui API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -168,7 +168,8 @@ el.replaceWith(d);
                 if state[&#39;context_mode&#39;] == self.mode_single:
                     active_raw = raw_bytes[start:end]
                 else:
-                    active_raw = raw_bytes[start:]
+                    # Concat only the bytes of valid blocks to skip intermediate empty/cancelled prompt noise
+                    active_raw = b&#34;&#34;.join(raw_bytes[b[0]:b[1]] for b in blocks[idx:])
                 return preview + &#34;\n&#34; + log_cleaner(active_raw.decode(errors=&#39;replace&#39;))
 
             def get_prompt_text():
@@ -207,7 +208,6 @@ el.replaceWith(d);
                     base_str = f&#39;\u25b6 Ctrl+\u2191/\u2193 adjusts by 50 lines  [Tab: {m_label}]&#39;
                 else:
                     idx = max(0, state[&#39;total_cmds&#39;] - state[&#39;context_cmd&#39;])
-                    import re
                     
                     def clean_preview(text):
                         # Limpia saltos de línea y el prompt inicial (todo hasta #, &gt; o $) para que quede solo el comando
@@ -370,10 +370,10 @@ el.replaceWith(d);
                 persona_title = &#34;Network Architect&#34; if active_persona == &#34;architect&#34; else &#34;Network Engineer&#34;
                 
                 active_buffer = get_active_buffer()
+                
                 live_text = &#34;&#34;
                 first_chunk = True
                 
-                import sys
                 from rich.rule import Rule
                 from rich.status import Status
                 from connpy.printer import IncrementalMarkdownParser
@@ -622,7 +622,8 @@ el.replaceWith(d);
             if state[&#39;context_mode&#39;] == self.mode_single:
                 active_raw = raw_bytes[start:end]
             else:
-                active_raw = raw_bytes[start:]
+                # Concat only the bytes of valid blocks to skip intermediate empty/cancelled prompt noise
+                active_raw = b&#34;&#34;.join(raw_bytes[b[0]:b[1]] for b in blocks[idx:])
             return preview + &#34;\n&#34; + log_cleaner(active_raw.decode(errors=&#39;replace&#39;))
 
         def get_prompt_text():
@@ -661,7 +662,6 @@ el.replaceWith(d);
                 base_str = f&#39;\u25b6 Ctrl+\u2191/\u2193 adjusts by 50 lines  [Tab: {m_label}]&#39;
             else:
                 idx = max(0, state[&#39;total_cmds&#39;] - state[&#39;context_cmd&#39;])
-                import re
                 
                 def clean_preview(text):
                     # Limpia saltos de línea y el prompt inicial (todo hasta #, &gt; o $) para que quede solo el comando
@@ -824,10 +824,10 @@ el.replaceWith(d);
             persona_title = &#34;Network Architect&#34; if active_persona == &#34;architect&#34; else &#34;Network Engineer&#34;
             
             active_buffer = get_active_buffer()
+            
             live_text = &#34;&#34;
             first_chunk = True
             
-            import sys
             from rich.rule import Rule
             from rich.status import Status
             from connpy.printer import IncrementalMarkdownParser
@@ -1017,7 +1017,7 @@ on_ai_call: async function(active_buffer, question) -&gt; result_dict</p></div>
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/cli/validators.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.cli.validators API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -508,7 +508,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/grpc_layer/connpy_pb2.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.grpc_layer.connpy_pb2 API documentation</title>
 <meta name="description" content="Generated protocol buffer code.">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -45,617 +45,6 @@ el.replaceWith(d);
 <section>
 </section>
 <section>
-<h2 class="section-title" id="header-classes">Classes</h2>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.AIResponse"><code class="flex name class">
-<span>class <span class="ident">AIResponse</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.AIResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.AskRequest"><code class="flex name class">
-<span>class <span class="ident">AskRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.AskRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.BoolResponse"><code class="flex name class">
-<span>class <span class="ident">BoolResponse</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.BoolResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.BulkRequest"><code class="flex name class">
-<span>class <span class="ident">BulkRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.BulkRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.CopilotRequest"><code class="flex name class">
-<span>class <span class="ident">CopilotRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.CopilotRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.CopilotResponse"><code class="flex name class">
-<span>class <span class="ident">CopilotResponse</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.CopilotResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.DeleteRequest"><code class="flex name class">
-<span>class <span class="ident">DeleteRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.DeleteRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.ExportRequest"><code class="flex name class">
-<span>class <span class="ident">ExportRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.ExportRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.FilterRequest"><code class="flex name class">
-<span>class <span class="ident">FilterRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.FilterRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.FullReplaceRequest"><code class="flex name class">
-<span>class <span class="ident">FullReplaceRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.FullReplaceRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.IdRequest"><code class="flex name class">
-<span>class <span class="ident">IdRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.IdRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.IntRequest"><code class="flex name class">
-<span>class <span class="ident">IntRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.IntRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.InteractRequest"><code class="flex name class">
-<span>class <span class="ident">InteractRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.InteractRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.InteractResponse"><code class="flex name class">
-<span>class <span class="ident">InteractResponse</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.InteractResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.ListRequest"><code class="flex name class">
-<span>class <span class="ident">ListRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.ListRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.MCPRequest"><code class="flex name class">
-<span>class <span class="ident">MCPRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.MCPRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.MessageValue"><code class="flex name class">
-<span>class <span class="ident">MessageValue</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.MessageValue.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.MoveRequest"><code class="flex name class">
-<span>class <span class="ident">MoveRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.MoveRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.NodeRequest"><code class="flex name class">
-<span>class <span class="ident">NodeRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.NodeRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.NodeRunResult"><code class="flex name class">
-<span>class <span class="ident">NodeRunResult</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.NodeRunResult.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.PluginRequest"><code class="flex name class">
-<span>class <span class="ident">PluginRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.PluginRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.ProfileRequest"><code class="flex name class">
-<span>class <span class="ident">ProfileRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.ProfileRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.ProviderRequest"><code class="flex name class">
-<span>class <span class="ident">ProviderRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.ProviderRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.RunRequest"><code class="flex name class">
-<span>class <span class="ident">RunRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.RunRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.ScriptRequest"><code class="flex name class">
-<span>class <span class="ident">ScriptRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.ScriptRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.StringRequest"><code class="flex name class">
-<span>class <span class="ident">StringRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.StringRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.StringResponse"><code class="flex name class">
-<span>class <span class="ident">StringResponse</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.StringResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.StructRequest"><code class="flex name class">
-<span>class <span class="ident">StructRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.StructRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.StructResponse"><code class="flex name class">
-<span>class <span class="ident">StructResponse</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.StructResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.TestRequest"><code class="flex name class">
-<span>class <span class="ident">TestRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.TestRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.UpdateRequest"><code class="flex name class">
-<span>class <span class="ident">UpdateRequest</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.UpdateRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-<dt id="connpy.grpc_layer.connpy_pb2.ValueResponse"><code class="flex name class">
-<span>class <span class="ident">ValueResponse</span></span>
-<span>(</span><span>*args, **kwargs)</span>
-</code></dt>
-<dd>
-<div class="desc"><p>A ProtocolMessage</p></div>
-<h3>Ancestors</h3>
-<ul class="hlist">
-<li>google._upb._message.Message</li>
-<li>google.protobuf.message.Message</li>
-</ul>
-<h3>Class variables</h3>
-<dl>
-<dt id="connpy.grpc_layer.connpy_pb2.ValueResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
-</dl>
-</dd>
-</dl>
 </section>
 </article>
 <nav id="sidebar">
@@ -668,207 +57,11 @@ el.replaceWith(d);
 <li><code><a title="connpy.grpc_layer" href="index.html">connpy.grpc_layer</a></code></li>
 </ul>
 </li>
-<li><h3><a href="#header-classes">Classes</a></h3>
-<ul>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.AIResponse" href="#connpy.grpc_layer.connpy_pb2.AIResponse">AIResponse</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.AIResponse.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.AIResponse.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.AskRequest" href="#connpy.grpc_layer.connpy_pb2.AskRequest">AskRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.AskRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.AskRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.BoolResponse" href="#connpy.grpc_layer.connpy_pb2.BoolResponse">BoolResponse</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.BoolResponse.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.BoolResponse.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.BulkRequest" href="#connpy.grpc_layer.connpy_pb2.BulkRequest">BulkRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.BulkRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.BulkRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.CopilotRequest" href="#connpy.grpc_layer.connpy_pb2.CopilotRequest">CopilotRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.CopilotRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.CopilotRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.CopilotResponse" href="#connpy.grpc_layer.connpy_pb2.CopilotResponse">CopilotResponse</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.CopilotResponse.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.CopilotResponse.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.DeleteRequest" href="#connpy.grpc_layer.connpy_pb2.DeleteRequest">DeleteRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.DeleteRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.DeleteRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.ExportRequest" href="#connpy.grpc_layer.connpy_pb2.ExportRequest">ExportRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.ExportRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.ExportRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.FilterRequest" href="#connpy.grpc_layer.connpy_pb2.FilterRequest">FilterRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.FilterRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.FilterRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.FullReplaceRequest" href="#connpy.grpc_layer.connpy_pb2.FullReplaceRequest">FullReplaceRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.FullReplaceRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.FullReplaceRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.IdRequest" href="#connpy.grpc_layer.connpy_pb2.IdRequest">IdRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.IdRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.IdRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.IntRequest" href="#connpy.grpc_layer.connpy_pb2.IntRequest">IntRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.IntRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.IntRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.InteractRequest" href="#connpy.grpc_layer.connpy_pb2.InteractRequest">InteractRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.InteractRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.InteractRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.InteractResponse" href="#connpy.grpc_layer.connpy_pb2.InteractResponse">InteractResponse</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.InteractResponse.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.InteractResponse.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.ListRequest" href="#connpy.grpc_layer.connpy_pb2.ListRequest">ListRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.ListRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.ListRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.MCPRequest" href="#connpy.grpc_layer.connpy_pb2.MCPRequest">MCPRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.MCPRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.MCPRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.MessageValue" href="#connpy.grpc_layer.connpy_pb2.MessageValue">MessageValue</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.MessageValue.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.MessageValue.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.MoveRequest" href="#connpy.grpc_layer.connpy_pb2.MoveRequest">MoveRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.MoveRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.MoveRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.NodeRequest" href="#connpy.grpc_layer.connpy_pb2.NodeRequest">NodeRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.NodeRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.NodeRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.NodeRunResult" href="#connpy.grpc_layer.connpy_pb2.NodeRunResult">NodeRunResult</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.NodeRunResult.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.NodeRunResult.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.PluginRequest" href="#connpy.grpc_layer.connpy_pb2.PluginRequest">PluginRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.PluginRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.PluginRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.ProfileRequest" href="#connpy.grpc_layer.connpy_pb2.ProfileRequest">ProfileRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.ProfileRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.ProfileRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.ProviderRequest" href="#connpy.grpc_layer.connpy_pb2.ProviderRequest">ProviderRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.ProviderRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.ProviderRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.RunRequest" href="#connpy.grpc_layer.connpy_pb2.RunRequest">RunRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.RunRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.RunRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.ScriptRequest" href="#connpy.grpc_layer.connpy_pb2.ScriptRequest">ScriptRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.ScriptRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.ScriptRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.StringRequest" href="#connpy.grpc_layer.connpy_pb2.StringRequest">StringRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.StringRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.StringRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.StringResponse" href="#connpy.grpc_layer.connpy_pb2.StringResponse">StringResponse</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.StringResponse.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.StringResponse.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.StructRequest" href="#connpy.grpc_layer.connpy_pb2.StructRequest">StructRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.StructRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.StructRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.StructResponse" href="#connpy.grpc_layer.connpy_pb2.StructResponse">StructResponse</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.StructResponse.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.StructResponse.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.TestRequest" href="#connpy.grpc_layer.connpy_pb2.TestRequest">TestRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.TestRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.TestRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.UpdateRequest" href="#connpy.grpc_layer.connpy_pb2.UpdateRequest">UpdateRequest</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.UpdateRequest.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.UpdateRequest.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-<li>
-<h4><code><a title="connpy.grpc_layer.connpy_pb2.ValueResponse" href="#connpy.grpc_layer.connpy_pb2.ValueResponse">ValueResponse</a></code></h4>
-<ul class="">
-<li><code><a title="connpy.grpc_layer.connpy_pb2.ValueResponse.DESCRIPTOR" href="#connpy.grpc_layer.connpy_pb2.ValueResponse.DESCRIPTOR">DESCRIPTOR</a></code></li>
-</ul>
-</li>
-</ul>
-</li>
 </ul>
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/grpc_layer/index.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.grpc_layer API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -102,7 +102,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/grpc_layer/remote_plugin_pb2.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.grpc_layer.remote_plugin_pb2 API documentation</title>
 <meta name="description" content="Generated protocol buffer code.">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -62,7 +62,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.remote_plugin_pb2.IdRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"></div>
+<div class="desc"><p>The type of the None singleton.</p></div>
 </dd>
 </dl>
 </dd>
@@ -81,7 +81,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.remote_plugin_pb2.OutputChunk.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"></div>
+<div class="desc"><p>The type of the None singleton.</p></div>
 </dd>
 </dl>
 </dd>
@@ -100,7 +100,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.remote_plugin_pb2.PluginInvokeRequest.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"></div>
+<div class="desc"><p>The type of the None singleton.</p></div>
 </dd>
 </dl>
 </dd>
@@ -119,7 +119,7 @@ el.replaceWith(d);
 <dl>
 <dt id="connpy.grpc_layer.remote_plugin_pb2.StringResponse.DESCRIPTOR"><code class="name">var <span class="ident">DESCRIPTOR</span></code></dt>
 <dd>
-<div class="desc"></div>
+<div class="desc"><p>The type of the None singleton.</p></div>
 </dd>
 </dl>
 </dd>
@@ -168,7 +168,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/grpc_layer/remote_plugin_pb2_grpc.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.grpc_layer.remote_plugin_pb2_grpc API documentation</title>
 <meta name="description" content="Client and server classes corresponding to protobuf-defined services.">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -366,7 +366,7 @@ def invoke_plugin(request,
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/grpc_layer/server.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.grpc_layer.server API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -177,9 +177,11 @@ el.replaceWith(d);
                 print(f&#34;AI Task Error: {e}&#34;)
                 traceback.print_exc()
                 chunk_queue.put((&#34;status&#34;, f&#34;Error: {str(e)}&#34;))
+                # Crucial: always send final_mark to avoid client deadlock
+                chunk_queue.put((&#34;final_mark&#34;, {&#34;response&#34;: f&#34;Error: {str(e)}&#34;, &#34;chat_history&#34;: history, &#34;error&#34;: True}))
 
         def request_listener():
-            nonlocal bridge, is_web, ai_thread, agent_instance
+            nonlocal bridge, is_web, ai_thread, agent_instance, history
             try:
                 for req in request_iterator:
                     if req.interrupt:
@@ -193,12 +195,21 @@ el.replaceWith(d);
 
                     if req.input_text:
                         is_web = &#34;web&#34; in (req.session_id or &#34;&#34;).lower() or (req.session_id or &#34;&#34;).lower().startswith(&#34;ws-&#34;)
+                        
+                        # Hydrate history from client if it&#39;s the first interaction in this stream
+                        if not history and req.chat_history:
+                            from .utils import from_value
+                            history = from_value(req.chat_history) or []
                         if not bridge:
                             bridge = StatusBridge(chunk_queue, request_queue=request_queue, is_web=is_web)
 
                         overrides = {}
                         if req.engineer_model: overrides[&#34;engineer_model&#34;] = req.engineer_model
                         if req.engineer_api_key: overrides[&#34;engineer_api_key&#34;] = req.engineer_api_key
+                        if req.architect_model: overrides[&#34;architect_model&#34;] = req.architect_model
+                        if req.architect_api_key: overrides[&#34;architect_api_key&#34;] = req.architect_api_key
+                        if req.HasField(&#34;engineer_auth&#34;): overrides[&#34;engineer_auth&#34;] = from_struct(req.engineer_auth)
+                        if req.HasField(&#34;architect_auth&#34;): overrides[&#34;architect_auth&#34;] = from_struct(req.architect_auth)
                         
                         # Start AI in its own thread so we can keep listening for interrupts
                         ai_thread = threading.Thread(
@@ -263,7 +274,8 @@ el.replaceWith(d);
 
     @handle_errors
     def list_sessions(self, request, context):
-        return connpy_pb2.ValueResponse(data=to_value(self.service.list_sessions()))
+        sessions, total = self.service.list_sessions()
+        return connpy_pb2.ValueResponse(data=to_value(sessions))
 
     @handle_errors
     def delete_session(self, request, context):
@@ -272,7 +284,8 @@ el.replaceWith(d);
 
     @handle_errors
     def configure_provider(self, request, context):
-        self.service.configure_provider(request.provider, request.model, request.api_key)
+        auth_dict = from_struct(request.auth) if request.HasField(&#34;auth&#34;) else None
+        self.service.configure_provider(request.provider, request.model, request.api_key, auth=auth_dict)
         return Empty()
     
     @handle_errors
@@ -286,6 +299,11 @@ el.replaceWith(d);
         )
         return Empty()
 
+    @handle_errors
+    def list_mcp_servers(self, request, context):
+        mcp_servers = self.service.list_mcp_servers()
+        return connpy_pb2.ValueResponse(data=to_value(mcp_servers))
+
     @handle_errors
     def load_session_data(self, request, context):
         return connpy_pb2.StructResponse(data=to_struct(self.service.load_session_data(request.value)))</code></pre>
@@ -305,6 +323,7 @@ el.replaceWith(d);
 <li><code><a title="connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.configure_provider" href="connpy_pb2_grpc.html#connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.configure_provider">configure_provider</a></code></li>
 <li><code><a title="connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.confirm" href="connpy_pb2_grpc.html#connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.confirm">confirm</a></code></li>
 <li><code><a title="connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.delete_session" href="connpy_pb2_grpc.html#connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.delete_session">delete_session</a></code></li>
+<li><code><a title="connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.list_mcp_servers" href="connpy_pb2_grpc.html#connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.list_mcp_servers">list_mcp_servers</a></code></li>
 <li><code><a title="connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.list_sessions" href="connpy_pb2_grpc.html#connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.list_sessions">list_sessions</a></code></li>
 <li><code><a title="connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.load_session_data" href="connpy_pb2_grpc.html#connpy.grpc_layer.connpy_pb2_grpc.AIServiceServicer.load_session_data">load_session_data</a></code></li>
 </ul>
@@ -975,7 +994,9 @@ interceptor chooses to service this RPC, or None otherwise.</p></div>
 
                 asyncio.run(n._async_interact_loop(remote_stream, resize_callback, copilot_handler=remote_copilot_handler))
             except Exception as e:
-                pass
+                import traceback
+                print(f&#34;[ERROR in run_async_loop] {e}&#34;)
+                traceback.print_exc()
             finally:
                 n._teardown_interact_environment()
                 response_queue.put(None)  # Signal EOF
@@ -1195,6 +1216,7 @@ interceptor chooses to service this RPC, or None otherwise.</p></div>
         self.service = ProfileService(config)
         self.node_service = NodeService(config)
 
+
     @handle_errors
     def list_profiles(self, request, context):
         f = request.filter_str if request.filter_str else None
@@ -1261,6 +1283,7 @@ interceptor chooses to service this RPC, or None otherwise.</p></div>
         self.on_interrupt = self._force_interrupt
         self.thread = None
         self.is_web = is_web
+        self.is_remote = True
 
     def _force_interrupt(self):
         &#34;&#34;&#34;Forcefully raise KeyboardInterrupt in the target thread.&#34;&#34;&#34;
@@ -1560,7 +1583,7 @@ interceptor chooses to service this RPC, or None otherwise.</p></div>
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/grpc_layer/stubs.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.grpc_layer.stubs API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -122,6 +122,10 @@ el.replaceWith(d);
         )
         if chat_history is not None:
             initial_req.chat_history.CopyFrom(to_value(chat_history))
+        if &#34;engineer_auth&#34; in overrides and overrides[&#34;engineer_auth&#34;]:
+            initial_req.engineer_auth.CopyFrom(to_struct(overrides[&#34;engineer_auth&#34;]))
+        if &#34;architect_auth&#34; in overrides and overrides[&#34;architect_auth&#34;]:
+            initial_req.architect_auth.CopyFrom(to_struct(overrides[&#34;architect_auth&#34;]))
             
         req_queue.put(initial_req)
 
@@ -135,6 +139,7 @@ el.replaceWith(d);
         
         full_content = &#34;&#34;
         header_printed = False
+        current_responder = &#34;engineer&#34;
         final_result = {&#34;response&#34;: &#34;&#34;, &#34;chat_history&#34;: []}
 
         # Background thread to pull responses from gRPC into a local queue
@@ -179,6 +184,10 @@ el.replaceWith(d);
                     break
 
                 if response.status_update:
+                    if response.status_update.startswith(&#34;__RESPONDER__:&#34;):
+                        current_responder = response.status_update.split(&#34;:&#34;)[1].lower()
+                        continue
+                        
                     if response.requires_confirmation:
                         if status: status.stop()
                         
@@ -231,7 +240,9 @@ el.replaceWith(d);
                             stable_console = RichConsole(theme=connpy_theme, file=get_original_stdout())
                             
                             # Print header on first chunk
-                            stable_console.print(Rule(&#34;[bold engineer]Network Engineer[/bold engineer]&#34;, style=&#34;engineer&#34;))
+                            alias = &#34;architect&#34; if current_responder == &#34;architect&#34; else &#34;engineer&#34;
+                            role_label = &#34;Network Architect&#34; if current_responder == &#34;architect&#34; else &#34;Network Engineer&#34;
+                            stable_console.print(Rule(f&#34;[bold {alias}]{role_label}[/bold {alias}]&#34;, style=alias))
                             header_printed = True
                             
                             # Initialize parser
@@ -283,16 +294,23 @@ el.replaceWith(d);
         return self.stub.confirm(connpy_pb2.StringRequest(value=input_text)).value
 
     @handle_errors
-    def list_sessions(self):
+    def list_sessions(self, limit=None):
-        return from_value(self.stub.list_sessions(Empty()).data)
+        from .utils import from_value
+        res = self.stub.list_sessions(Empty())
+        sessions = from_value(res.data) or []
+        if limit and len(sessions) &gt; limit:
+            return sessions[:limit], len(sessions)
+        return sessions, len(sessions)
 
     @handle_errors
     def delete_session(self, session_id):
         self.stub.delete_session(connpy_pb2.StringRequest(value=session_id))
 
     @handle_errors
-    def configure_provider(self, provider, model=None, api_key=None):
+    def configure_provider(self, provider, model=None, api_key=None, auth=None):
         req = connpy_pb2.ProviderRequest(provider=provider, model=model or &#34;&#34;, api_key=api_key or &#34;&#34;)
+        if auth:
+            req.auth.CopyFrom(to_struct(auth))
         self.stub.configure_provider(req)
 
     @handle_errors
@@ -306,6 +324,11 @@ el.replaceWith(d);
         )
         self.stub.configure_mcp(req)
 
+    @handle_errors
+    def list_mcp_servers(self):
+        res = self.stub.list_mcp_servers(Empty())
+        return from_value(res.data) or {}
+
     @handle_errors
     def load_session_data(self, session_id):
         return from_struct(self.stub.load_session_data(connpy_pb2.StringRequest(value=session_id)).data)</code></pre>
@@ -344,6 +367,10 @@ def ask(self, input_text, dryrun=False, chat_history=None, session_id=None, debu
     )
     if chat_history is not None:
         initial_req.chat_history.CopyFrom(to_value(chat_history))
+    if &#34;engineer_auth&#34; in overrides and overrides[&#34;engineer_auth&#34;]:
+        initial_req.engineer_auth.CopyFrom(to_struct(overrides[&#34;engineer_auth&#34;]))
+    if &#34;architect_auth&#34; in overrides and overrides[&#34;architect_auth&#34;]:
+        initial_req.architect_auth.CopyFrom(to_struct(overrides[&#34;architect_auth&#34;]))
         
     req_queue.put(initial_req)
 
@@ -357,6 +384,7 @@ def ask(self, input_text, dryrun=False, chat_history=None, session_id=None, debu
     
     full_content = &#34;&#34;
     header_printed = False
+    current_responder = &#34;engineer&#34;
     final_result = {&#34;response&#34;: &#34;&#34;, &#34;chat_history&#34;: []}
 
     # Background thread to pull responses from gRPC into a local queue
@@ -401,6 +429,10 @@ def ask(self, input_text, dryrun=False, chat_history=None, session_id=None, debu
                 break
 
             if response.status_update:
+                if response.status_update.startswith(&#34;__RESPONDER__:&#34;):
+                    current_responder = response.status_update.split(&#34;:&#34;)[1].lower()
+                    continue
+                    
                 if response.requires_confirmation:
                     if status: status.stop()
                     
@@ -453,7 +485,9 @@ def ask(self, input_text, dryrun=False, chat_history=None, session_id=None, debu
                         stable_console = RichConsole(theme=connpy_theme, file=get_original_stdout())
                         
                         # Print header on first chunk
-                        stable_console.print(Rule(&#34;[bold engineer]Network Engineer[/bold engineer]&#34;, style=&#34;engineer&#34;))
+                        alias = &#34;architect&#34; if current_responder == &#34;architect&#34; else &#34;engineer&#34;
+                        role_label = &#34;Network Architect&#34; if current_responder == &#34;architect&#34; else &#34;Network Engineer&#34;
+                        stable_console.print(Rule(f&#34;[bold {alias}]{role_label}[/bold {alias}]&#34;, style=alias))
                         header_printed = True
                         
                         # Initialize parser
@@ -524,7 +558,7 @@ def configure_mcp(self, name, url=None, enabled=True, auto_load_on_os=None, remo
 <div class="desc"></div>
 </dd>
 <dt id="connpy.grpc_layer.stubs.AIStub.configure_provider"><code class="name flex">
-<span>def <span class="ident">configure_provider</span></span>(<span>self, provider, model=None, api_key=None)</span>
+<span>def <span class="ident">configure_provider</span></span>(<span>self, provider, model=None, api_key=None, auth=None)</span>
 </code></dt>
 <dd>
 <details class="source">
@@ -532,8 +566,10 @@ def configure_mcp(self, name, url=None, enabled=True, auto_load_on_os=None, remo
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">@handle_errors
-def configure_provider(self, provider, model=None, api_key=None):
+def configure_provider(self, provider, model=None, api_key=None, auth=None):
     req = connpy_pb2.ProviderRequest(provider=provider, model=model or &#34;&#34;, api_key=api_key or &#34;&#34;)
+    if auth:
+        req.auth.CopyFrom(to_struct(auth))
     self.stub.configure_provider(req)</code></pre>
 </details>
 <div class="desc"></div>
@@ -566,8 +602,8 @@ def delete_session(self, session_id):
 </details>
 <div class="desc"></div>
 </dd>
-<dt id="connpy.grpc_layer.stubs.AIStub.list_sessions"><code class="name flex">
+<dt id="connpy.grpc_layer.stubs.AIStub.list_mcp_servers"><code class="name flex">
-<span>def <span class="ident">list_sessions</span></span>(<span>self)</span>
+<span>def <span class="ident">list_mcp_servers</span></span>(<span>self)</span>
 </code></dt>
 <dd>
 <details class="source">
@@ -575,8 +611,28 @@ def delete_session(self, session_id):
 <span>Expand source code</span>
 </summary>
 <pre><code class="python">@handle_errors
-def list_sessions(self):
+def list_mcp_servers(self):
-    return from_value(self.stub.list_sessions(Empty()).data)</code></pre>
+    res = self.stub.list_mcp_servers(Empty())
+    return from_value(res.data) or {}</code></pre>
+</details>
+<div class="desc"></div>
+</dd>
+<dt id="connpy.grpc_layer.stubs.AIStub.list_sessions"><code class="name flex">
+<span>def <span class="ident">list_sessions</span></span>(<span>self, limit=None)</span>
+</code></dt>
+<dd>
+<details class="source">
+<summary>
+<span>Expand source code</span>
+</summary>
+<pre><code class="python">@handle_errors
+def list_sessions(self, limit=None):
+    from .utils import from_value
+    res = self.stub.list_sessions(Empty())
+    sessions = from_value(res.data) or []
+    if limit and len(sessions) &gt; limit:
+        return sessions[:limit], len(sessions)
+    return sessions, len(sessions)</code></pre>
 </details>
 <div class="desc"></div>
 </dd>
@@ -2470,6 +2526,7 @@ def stop_api(self):
 <li><code><a title="connpy.grpc_layer.stubs.AIStub.configure_provider" href="#connpy.grpc_layer.stubs.AIStub.configure_provider">configure_provider</a></code></li>
 <li><code><a title="connpy.grpc_layer.stubs.AIStub.confirm" href="#connpy.grpc_layer.stubs.AIStub.confirm">confirm</a></code></li>
 <li><code><a title="connpy.grpc_layer.stubs.AIStub.delete_session" href="#connpy.grpc_layer.stubs.AIStub.delete_session">delete_session</a></code></li>
+<li><code><a title="connpy.grpc_layer.stubs.AIStub.list_mcp_servers" href="#connpy.grpc_layer.stubs.AIStub.list_mcp_servers">list_mcp_servers</a></code></li>
 <li><code><a title="connpy.grpc_layer.stubs.AIStub.list_sessions" href="#connpy.grpc_layer.stubs.AIStub.list_sessions">list_sessions</a></code></li>
 <li><code><a title="connpy.grpc_layer.stubs.AIStub.load_session_data" href="#connpy.grpc_layer.stubs.AIStub.load_session_data">load_session_data</a></code></li>
 </ul>
@@ -2561,7 +2618,7 @@ def stop_api(self):
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/grpc_layer/utils.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.grpc_layer.utils API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -138,7 +138,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/index.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy API documentation</title>
 <meta name="description" content="&lt;p align=&#34;center&#34;&gt;
 &lt;img src=&#34;https://nginx.gederico.dynu.net/images/CONNPY-resized.png&#34; alt=&#34;App Logo&#34;&gt;
@@ -205,10 +205,6 @@ response = myai.ask(&quot;What is the status of the BGP neighbors in the office?
 <dd>
 <div class="desc"></div>
 </dd>
-<dt><code class="name"><a title="connpy.tests" href="tests/index.html">connpy.tests</a></code></dt>
-<dd>
-<div class="desc"></div>
-</dd>
 <dt><code class="name"><a title="connpy.tunnels" href="tunnels.html">connpy.tunnels</a></code></dt>
 <dd>
 <div class="desc"></div>
@@ -620,7 +616,7 @@ indicating successful verification.</p>
 </dd>
 <dt id="connpy.ai"><code class="flex name class">
 <span>class <span class="ident">ai</span></span>
-<span>(</span><span>config,<br>org=None,<br>api_key=None,<br>engineer_model=None,<br>architect_model=None,<br>engineer_api_key=None,<br>architect_api_key=None,<br>console=None,<br>confirm_handler=None,<br>trust=False)</span>
+<span>(</span><span>config,<br>org=None,<br>api_key=None,<br>engineer_model=None,<br>architect_model=None,<br>engineer_api_key=None,<br>architect_api_key=None,<br>console=None,<br>confirm_handler=None,<br>trust=False,<br>engineer_auth=None,<br>architect_auth=None,<br>**kwargs)</span>
 </code></dt>
 <dd>
 <details class="source">
@@ -638,7 +634,7 @@ class ai:
         r&#39;^systemctl\s+status\s+&#39;, r&#39;^journalctl\s+&#39;
     ]
 
-    def __init__(self, config, org=None, api_key=None, engineer_model=None, architect_model=None, engineer_api_key=None, architect_api_key=None, console=None, confirm_handler=None, trust=False):
+    def __init__(self, config, org=None, api_key=None, engineer_model=None, architect_model=None, engineer_api_key=None, architect_api_key=None, console=None, confirm_handler=None, trust=False, engineer_auth=None, architect_auth=None, **kwargs):
         self.config = config
         self.console = console or printer.console
         self.confirm_handler = confirm_handler or self._local_confirm_handler
@@ -657,6 +653,29 @@ class ai:
         self.engineer_key = engineer_api_key or aiconfig.get(&#34;engineer_api_key&#34;)
         self.architect_key = architect_api_key or aiconfig.get(&#34;architect_api_key&#34;)
 
+        # Auth configurations (Prioridad: Argumento -&gt; Config)
+        self.engineer_auth = engineer_auth if engineer_auth is not None else aiconfig.get(&#34;engineer_auth&#34;)
+        if self.engineer_auth is None:
+            self.engineer_auth = {}
+        elif not isinstance(self.engineer_auth, dict):
+            self.engineer_auth = {}
+
+        self.architect_auth = architect_auth if architect_auth is not None else aiconfig.get(&#34;architect_auth&#34;)
+        if self.architect_auth is None:
+            self.architect_auth = {}
+        elif not isinstance(self.architect_auth, dict):
+            self.architect_auth = {}
+
+        # Backward compatibility fallbacks: only inject api_key if the auth dict is empty/not configured
+        if self.engineer_key and not self.engineer_auth:
+            self.engineer_auth[&#34;api_key&#34;] = self.engineer_key
+        if self.architect_key and not self.architect_auth:
+            self.architect_auth[&#34;api_key&#34;] = self.architect_key
+
+        # Strategic Reasoning Engine (Architect) availability
+        is_architect_keyless = &#34;vertex&#34; in self.architect_model.lower() or &#34;ollama&#34; in self.architect_model.lower() or &#34;local&#34; in self.architect_model.lower()
+        self.has_architect = bool(self.architect_key or self.architect_auth or is_architect_keyless)
+
         # Custom Trusted Commands Regexes
         custom_trusted = aiconfig.get(&#34;trusted_commands&#34;, [])
         if isinstance(custom_trusted, str):
@@ -697,12 +716,12 @@ class ai:
         # Session Management
         self.sessions_dir = os.path.join(self.config.defaultdir, &#34;ai_sessions&#34;)
         os.makedirs(self.sessions_dir, exist_ok=True)
-        self.session_id = None
+        self.session_id = getattr(self.config, &#34;session_id&#34;, None)
-        self.session_path = None
+        self.session_path = os.path.join(self.sessions_dir, f&#34;{self.session_id}.json&#34;) if self.session_id else None
 
         # Prompts base agnósticos
         architect_instructions = &#34;&#34;
-        if self.architect_key:
+        if self.has_architect:
             architect_instructions = &#34;&#34;&#34;
             CRITICAL - CONSULT vs ESCALATE:
             - ALWAYS use &#39;consult_architect&#39; for: Configuration planning, design decisions, complex troubleshooting.
@@ -718,7 +737,7 @@ class ai:
         else:
             architect_instructions = &#34;&#34;&#34;
             CRITICAL - ARCHITECT UNAVAILABLE:
-            - The Strategic Reasoning Engine (Architect) is currently UNAVAILABLE because its API key is not configured.
+            - The Strategic Reasoning Engine (Architect) is currently UNAVAILABLE because its API key or authentication is not configured.
             - DO NOT attempt to consult or escalate to the architect.
             - If the user asks to consult the architect, inform them that the Architect is offline and offer to help them directly to the best of your abilities.
 &#34;&#34;&#34;
@@ -824,15 +843,19 @@ class ai:
         if status_formatter:
             self.tool_status_formatters[name] = status_formatter
 
-    def _stream_completion(self, model, messages, tools, api_key, status=None, label=&#34;&#34;, debug=False, chunk_callback=None, **kwargs):
+    def _stream_completion(self, model, messages, tools, api_key=None, status=None, label=&#34;&#34;, debug=False, chunk_callback=None, auth=None, **kwargs):
         &#34;&#34;&#34;Stream a completion call, rendering styled Markdown in real-time.
 
         Returns (response, streamed) where:
         - response: reconstructed ModelResponse (same as non-streaming)
         - streamed: True if text was rendered to console during streaming
         &#34;&#34;&#34;
+        auth_dict = auth if auth is not None else {}
+        if api_key and &#34;api_key&#34; not in auth_dict:
+            auth_dict = auth_dict.copy()
+            auth_dict[&#34;api_key&#34;] = api_key
 
-        stream_resp = completion(model=model, messages=messages, tools=tools, api_key=api_key, stream=True, **kwargs)
+        stream_resp = completion(model=model, messages=messages, tools=tools, stream=True, **auth_dict, **kwargs)
 
         chunks = []
         full_content = &#34;&#34;
@@ -1275,7 +1298,7 @@ class ai:
                 
                 try:
                     safe_messages = self._sanitize_messages(messages)
-                    response = completion(model=self.engineer_model, messages=safe_messages, tools=tools, api_key=self.engineer_key)
+                    response = completion(model=self.engineer_model, messages=safe_messages, tools=tools, **self.engineer_auth)
                 except Exception as e:
                     if status: status.stop()
                     raise ValueError(f&#34;Engineer failed to connect: {str(e)}&#34;)
@@ -1409,16 +1432,27 @@ class ai:
                     continue
         return sorted(sessions, key=lambda x: x[&#34;created_at&#34;], reverse=True)
 
-    def list_sessions(self):
+    def list_sessions(self, limit=20):
         &#34;&#34;&#34;Prints a list of sessions using printer.table.&#34;&#34;&#34;
         sessions = self._get_sessions()
         if not sessions:
             printer.info(&#34;No saved AI sessions found.&#34;)
             return
         
+        total = len(sessions)
+        if limit and total &gt; limit:
+            sessions = sessions[:limit]
+            
         columns = [&#34;ID&#34;, &#34;Title&#34;, &#34;Created At&#34;, &#34;Model&#34;]
         rows = [[s[&#34;id&#34;], s[&#34;title&#34;], s[&#34;created_at&#34;], s[&#34;model&#34;]] for s in sessions]
-        printer.table(&#34;AI Persisted Sessions&#34;, columns, rows)
+        
+        title = &#34;AI Persisted Sessions&#34;
+        if limit and total &gt; limit:
+            title += f&#34; (Showing last {limit} of {total})&#34;
+            
+        printer.table(title, columns, rows)
+        if limit and total &gt; limit:
+            printer.info(f&#34;Use &#39;--list --all&#39; (if supported) or check the sessions directory to see all {total} sessions.&#34;)
 
     def load_session_data(self, session_id):
         &#34;&#34;&#34;Loads a session&#39;s raw data by ID.&#34;&#34;&#34;
@@ -1449,8 +1483,10 @@ class ai:
         return sessions[0][&#34;id&#34;] if sessions else None
 
     def _generate_session_id(self, query):
-        &#34;&#34;&#34;Generates a unique session ID based on timestamp.&#34;&#34;&#34;
+        &#34;&#34;&#34;Generates a unique session ID based on timestamp and a random suffix.&#34;&#34;&#34;
-        return datetime.datetime.now().strftime(&#34;%Y%m%d-%H%M%S&#34;)
+        ts = datetime.datetime.now().strftime(&#34;%Y%m%d-%H%M%S&#34;)
+        suffix = secrets.token_hex(2)
+        return f&#34;{ts}-{suffix}&#34;
 
     def save_session(self, history, title=None, model=None):
         &#34;&#34;&#34;Saves current history to the session file.&#34;&#34;&#34;
@@ -1459,6 +1495,8 @@ class ai:
             first_user_msg = next((m[&#34;content&#34;] for m in history if m[&#34;role&#34;] == &#34;user&#34;), &#34;new-session&#34;)
             self.session_id = self._generate_session_id(first_user_msg)
             self.session_path = os.path.join(self.sessions_dir, f&#34;{self.session_id}.json&#34;)
+        elif not self.session_path:
+            self.session_path = os.path.join(self.sessions_dir, f&#34;{self.session_id}.json&#34;)
 
         # If it&#39;s a new file, we might want to set a better title
         if not os.path.exists(self.session_path) and not title:
@@ -1496,16 +1534,22 @@ class ai:
 
     @MethodHook
     def ask(self, user_input, dryrun=False, chat_history=None, status=None, debug=False, stream=True, session_id=None, chunk_callback=None):
-        if not self.engineer_key:
+        is_engineer_keyless = &#34;vertex&#34; in self.engineer_model.lower() or &#34;ollama&#34; in self.engineer_model.lower() or &#34;local&#34; in self.engineer_model.lower()
-            raise ValueError(&#34;Engineer API key not configured. Use &#39;connpy config --engineer-api-key &lt;key&gt;&#39; to set it.&#34;)
+        if not self.engineer_key and not self.engineer_auth and not is_engineer_keyless:
+            raise ValueError(&#34;Engineer API key or authentication not configured. Use &#39;connpy config --engineer-auth &lt;auth&gt;&#39; to set it.&#34;)
             
         if chat_history is None: chat_history = []
         
         # Load session if provided and history is empty
-        if session_id and not chat_history:
+        if session_id:
-            session_data = self.load_session_data(session_id)
+            # Force the session_id even if it doesn&#39;t exist yet
-            if session_data:
+            self.session_id = session_id
-                chat_history = session_data.get(&#34;history&#34;, [])
+            self.session_path = os.path.join(self.sessions_dir, f&#34;{session_id}.json&#34;)
+            
+            if not chat_history:
+                session_data = self.load_session_data(session_id)
+                if session_data:
+                    chat_history = session_data.get(&#34;history&#34;, [])
                 # If we loaded history, the caller might need it back
                 # But typically ask() is called in a loop with an external history object
 
@@ -1541,6 +1585,7 @@ class ai:
         tools = self._get_architect_tools() if current_brain == &#34;architect&#34; else self._get_engineer_tools()
         model = self.architect_model if current_brain == &#34;architect&#34; else self.engineer_model
         key = self.architect_key if current_brain == &#34;architect&#34; else self.engineer_key
+        current_auth = self.architect_auth if current_brain == &#34;architect&#34; else self.engineer_auth
 
         # Estructura optimizada para Prompt Caching (Solo para Anthropic directo, Vertex tiene reglas distintas)
         if &#34;claude&#34; in model.lower() and &#34;vertex&#34; not in model.lower():
@@ -1590,8 +1635,8 @@ class ai:
                 
                 label = &#34;[architect][bold]Architect[/bold][/architect]&#34; if current_brain == &#34;architect&#34; else &#34;[engineer][bold]Engineer[/bold][/engineer]&#34;
                 if status: 
-                    # Notify responder identity ONLY for web/remote clients (StatusBridge has is_web)
+                    # Notify responder identity for web/remote clients
-                    if getattr(status, &#34;is_web&#34;, False):
+                    if getattr(status, &#34;is_web&#34;, False) or getattr(status, &#34;is_remote&#34;, False):
                         status.update(f&#34;__RESPONDER__:{current_brain}&#34;)
                     status.update(f&#34;{label} is thinking... (step {iteration})&#34;)
                 
@@ -1600,12 +1645,12 @@ class ai:
                     safe_messages = self._sanitize_messages(messages)
                     if stream:
                         response, streamed_response = self._stream_completion(
-                            model=model, messages=safe_messages, tools=tools, api_key=key,
+                            model=model, messages=safe_messages, tools=tools, auth=current_auth,
                             status=status, label=label, debug=debug, num_retries=3,
                             chunk_callback=chunk_callback
                         )
                     else:
-                        response = completion(model=model, messages=safe_messages, tools=tools, api_key=key, num_retries=3)
+                        response = completion(model=model, messages=safe_messages, tools=tools, num_retries=3, **current_auth)
                 except Exception as e:
                     if current_brain == &#34;architect&#34;:
                         if status: status.update(&#34;[unavailable]Architect unavailable! Falling back to Engineer...&#34;)
@@ -1614,6 +1659,7 @@ class ai:
                         model = self.engineer_model
                         tools = self._get_engineer_tools()
                         key = self.engineer_key
+                        current_auth = self.engineer_auth
                         # Rebuild messages with Engineer system prompt and original user request
                         messages = [{&#34;role&#34;: &#34;system&#34;, &#34;content&#34;: self.engineer_system_prompt}]
                         # Add chat history if exists (excluding system prompt)
@@ -1706,6 +1752,7 @@ class ai:
                         model = self.architect_model
                         tools = self._get_architect_tools()
                         key = self.architect_key
+                        current_auth = self.architect_auth
                         messages[0] = {&#34;role&#34;: &#34;system&#34;, &#34;content&#34;: self.architect_system_prompt}
                         # Prepare handover context to inject AFTER all tool responses
                         handover_msg = f&#34;HANDOVER FROM EXECUTION ENGINE\n\nReason: {args[&#39;reason&#39;]}\n\nContext: {args[&#39;context&#39;]}\n\nYou are now in control of this conversation.&#34;
@@ -1727,6 +1774,7 @@ class ai:
                         model = self.engineer_model
                         tools = self._get_engineer_tools()
                         key = self.engineer_key
+                        current_auth = self.engineer_auth
                         messages[0] = {&#34;role&#34;: &#34;system&#34;, &#34;content&#34;: self.engineer_system_prompt}
                         # Prepare handover context to inject AFTER all tool responses
                         handover_msg = f&#34;HANDOVER FROM ARCHITECT\n\nSummary: {args[&#39;summary&#39;]}\n\nYou are now back in control. Continue handling the user&#39;s requests.&#34;
@@ -1768,7 +1816,7 @@ class ai:
                     messages.append({&#34;role&#34;: &#34;user&#34;, &#34;content&#34;: &#34;Hard iteration limit reached. Please provide a summary of your findings so far.&#34;})
                     try:
                         safe_messages = self._sanitize_messages(messages)
-                        response = completion(model=model, messages=safe_messages, tools=[], api_key=key)
+                        response = completion(model=model, messages=safe_messages, tools=[], **current_auth)
                         resp_msg = response.choices[0].message
                         messages.append(resp_msg.model_dump(exclude_none=True))
                     except Exception as e:
@@ -1788,7 +1836,7 @@ class ai:
             try:
                 safe_messages = self._sanitize_messages(summary_messages)
                 # Use tools=None to force a text summary during interruption
-                response = completion(model=model, messages=safe_messages, tools=None, api_key=key)
+                response = completion(model=model, messages=safe_messages, tools=None, **current_auth)
                 resp_msg = response.choices[0].message
                 messages.append(resp_msg.model_dump(exclude_none=True))
                 
@@ -1925,6 +1973,7 @@ Node: {node_name}&#34;&#34;&#34;
         # Use models based on persona
         current_model = self.architect_model if persona == &#34;architect&#34; else self.engineer_model
         current_key = self.architect_key if persona == &#34;architect&#34; else self.engineer_key
+        current_auth = self.architect_auth if persona == &#34;architect&#34; else self.engineer_auth
 
         try:
             while iteration &lt; max_iterations:
@@ -1934,8 +1983,8 @@ Node: {node_name}&#34;&#34;&#34;
                     model=current_model,
                     messages=messages,
                     tools=mcp_tools if mcp_tools else None,
-                    api_key=current_key,
+                    stream=True,
-                    stream=True
+                    **current_auth
                 )
                 
                 full_content = &#34;&#34;
@@ -2008,8 +2057,8 @@ Node: {node_name}&#34;&#34;&#34;
                     model=self.engineer_model,
                     messages=messages,
                     tools=None,
-                    api_key=self.engineer_key,
+                    stream=True,
-                    stream=True
+                    **self.engineer_auth
                 )
                 
                 full_content = &#34;&#34;
@@ -2095,7 +2144,7 @@ Node: {node_name}&#34;&#34;&#34;
 <dl>
 <dt id="connpy.ai.SAFE_COMMANDS"><code class="name">var <span class="ident">SAFE_COMMANDS</span></code></dt>
 <dd>
-<div class="desc"></div>
+<div class="desc"><p>The type of the None singleton.</p></div>
 </dd>
 </dl>
 <h3>Instance variables</h3>
@@ -2255,6 +2304,7 @@ Node: {node_name}&#34;&#34;&#34;
         # Use models based on persona
         current_model = self.architect_model if persona == &#34;architect&#34; else self.engineer_model
         current_key = self.architect_key if persona == &#34;architect&#34; else self.engineer_key
+        current_auth = self.architect_auth if persona == &#34;architect&#34; else self.engineer_auth
 
         try:
             while iteration &lt; max_iterations:
@@ -2264,8 +2314,8 @@ Node: {node_name}&#34;&#34;&#34;
                     model=current_model,
                     messages=messages,
                     tools=mcp_tools if mcp_tools else None,
-                    api_key=current_key,
+                    stream=True,
-                    stream=True
+                    **current_auth
                 )
                 
                 full_content = &#34;&#34;
@@ -2338,8 +2388,8 @@ Node: {node_name}&#34;&#34;&#34;
                     model=self.engineer_model,
                     messages=messages,
                     tools=None,
-                    api_key=self.engineer_key,
+                    stream=True,
-                    stream=True
+                    **self.engineer_auth
                 )
                 
                 full_content = &#34;&#34;
@@ -2429,16 +2479,22 @@ Node: {node_name}&#34;&#34;&#34;
 </summary>
 <pre><code class="python">@MethodHook
 def ask(self, user_input, dryrun=False, chat_history=None, status=None, debug=False, stream=True, session_id=None, chunk_callback=None):
-    if not self.engineer_key:
+    is_engineer_keyless = &#34;vertex&#34; in self.engineer_model.lower() or &#34;ollama&#34; in self.engineer_model.lower() or &#34;local&#34; in self.engineer_model.lower()
-        raise ValueError(&#34;Engineer API key not configured. Use &#39;connpy config --engineer-api-key &lt;key&gt;&#39; to set it.&#34;)
+    if not self.engineer_key and not self.engineer_auth and not is_engineer_keyless:
+        raise ValueError(&#34;Engineer API key or authentication not configured. Use &#39;connpy config --engineer-auth &lt;auth&gt;&#39; to set it.&#34;)
         
     if chat_history is None: chat_history = []
     
     # Load session if provided and history is empty
-    if session_id and not chat_history:
+    if session_id:
-        session_data = self.load_session_data(session_id)
+        # Force the session_id even if it doesn&#39;t exist yet
-        if session_data:
+        self.session_id = session_id
-            chat_history = session_data.get(&#34;history&#34;, [])
+        self.session_path = os.path.join(self.sessions_dir, f&#34;{session_id}.json&#34;)
+        
+        if not chat_history:
+            session_data = self.load_session_data(session_id)
+            if session_data:
+                chat_history = session_data.get(&#34;history&#34;, [])
             # If we loaded history, the caller might need it back
             # But typically ask() is called in a loop with an external history object
 
@@ -2474,6 +2530,7 @@ def ask(self, user_input, dryrun=False, chat_history=None, status=None, debug=Fa
     tools = self._get_architect_tools() if current_brain == &#34;architect&#34; else self._get_engineer_tools()
     model = self.architect_model if current_brain == &#34;architect&#34; else self.engineer_model
     key = self.architect_key if current_brain == &#34;architect&#34; else self.engineer_key
+    current_auth = self.architect_auth if current_brain == &#34;architect&#34; else self.engineer_auth
 
     # Estructura optimizada para Prompt Caching (Solo para Anthropic directo, Vertex tiene reglas distintas)
     if &#34;claude&#34; in model.lower() and &#34;vertex&#34; not in model.lower():
@@ -2523,8 +2580,8 @@ def ask(self, user_input, dryrun=False, chat_history=None, status=None, debug=Fa
             
             label = &#34;[architect][bold]Architect[/bold][/architect]&#34; if current_brain == &#34;architect&#34; else &#34;[engineer][bold]Engineer[/bold][/engineer]&#34;
             if status: 
-                # Notify responder identity ONLY for web/remote clients (StatusBridge has is_web)
+                # Notify responder identity for web/remote clients
-                if getattr(status, &#34;is_web&#34;, False):
+                if getattr(status, &#34;is_web&#34;, False) or getattr(status, &#34;is_remote&#34;, False):
                     status.update(f&#34;__RESPONDER__:{current_brain}&#34;)
                 status.update(f&#34;{label} is thinking... (step {iteration})&#34;)
             
@@ -2533,12 +2590,12 @@ def ask(self, user_input, dryrun=False, chat_history=None, status=None, debug=Fa
                 safe_messages = self._sanitize_messages(messages)
                 if stream:
                     response, streamed_response = self._stream_completion(
-                        model=model, messages=safe_messages, tools=tools, api_key=key,
+                        model=model, messages=safe_messages, tools=tools, auth=current_auth,
                         status=status, label=label, debug=debug, num_retries=3,
                         chunk_callback=chunk_callback
                     )
                 else:
-                    response = completion(model=model, messages=safe_messages, tools=tools, api_key=key, num_retries=3)
+                    response = completion(model=model, messages=safe_messages, tools=tools, num_retries=3, **current_auth)
             except Exception as e:
                 if current_brain == &#34;architect&#34;:
                     if status: status.update(&#34;[unavailable]Architect unavailable! Falling back to Engineer...&#34;)
@@ -2547,6 +2604,7 @@ def ask(self, user_input, dryrun=False, chat_history=None, status=None, debug=Fa
                     model = self.engineer_model
                     tools = self._get_engineer_tools()
                     key = self.engineer_key
+                    current_auth = self.engineer_auth
                     # Rebuild messages with Engineer system prompt and original user request
                     messages = [{&#34;role&#34;: &#34;system&#34;, &#34;content&#34;: self.engineer_system_prompt}]
                     # Add chat history if exists (excluding system prompt)
@@ -2639,6 +2697,7 @@ def ask(self, user_input, dryrun=False, chat_history=None, status=None, debug=Fa
                     model = self.architect_model
                     tools = self._get_architect_tools()
                     key = self.architect_key
+                    current_auth = self.architect_auth
                     messages[0] = {&#34;role&#34;: &#34;system&#34;, &#34;content&#34;: self.architect_system_prompt}
                     # Prepare handover context to inject AFTER all tool responses
                     handover_msg = f&#34;HANDOVER FROM EXECUTION ENGINE\n\nReason: {args[&#39;reason&#39;]}\n\nContext: {args[&#39;context&#39;]}\n\nYou are now in control of this conversation.&#34;
@@ -2660,6 +2719,7 @@ def ask(self, user_input, dryrun=False, chat_history=None, status=None, debug=Fa
                     model = self.engineer_model
                     tools = self._get_engineer_tools()
                     key = self.engineer_key
+                    current_auth = self.engineer_auth
                     messages[0] = {&#34;role&#34;: &#34;system&#34;, &#34;content&#34;: self.engineer_system_prompt}
                     # Prepare handover context to inject AFTER all tool responses
                     handover_msg = f&#34;HANDOVER FROM ARCHITECT\n\nSummary: {args[&#39;summary&#39;]}\n\nYou are now back in control. Continue handling the user&#39;s requests.&#34;
@@ -2701,7 +2761,7 @@ def ask(self, user_input, dryrun=False, chat_history=None, status=None, debug=Fa
                 messages.append({&#34;role&#34;: &#34;user&#34;, &#34;content&#34;: &#34;Hard iteration limit reached. Please provide a summary of your findings so far.&#34;})
                 try:
                     safe_messages = self._sanitize_messages(messages)
-                    response = completion(model=model, messages=safe_messages, tools=[], api_key=key)
+                    response = completion(model=model, messages=safe_messages, tools=[], **current_auth)
                     resp_msg = response.choices[0].message
                     messages.append(resp_msg.model_dump(exclude_none=True))
                 except Exception as e:
@@ -2721,7 +2781,7 @@ def ask(self, user_input, dryrun=False, chat_history=None, status=None, debug=Fa
         try:
             safe_messages = self._sanitize_messages(summary_messages)
             # Use tools=None to force a text summary during interruption
-            response = completion(model=model, messages=safe_messages, tools=None, api_key=key)
+            response = completion(model=model, messages=safe_messages, tools=None, **current_auth)
             resp_msg = response.choices[0].message
             messages.append(resp_msg.model_dump(exclude_none=True))
             
@@ -2844,23 +2904,34 @@ def confirm(self, user_input): return True</code></pre>
 <div class="desc"><p>List nodes matching the filter pattern. Returns metadata for &lt;=5 nodes, names only for more.</p></div>
 </dd>
 <dt id="connpy.ai.list_sessions"><code class="name flex">
-<span>def <span class="ident">list_sessions</span></span>(<span>self)</span>
+<span>def <span class="ident">list_sessions</span></span>(<span>self, limit=20)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
-<pre><code class="python">def list_sessions(self):
+<pre><code class="python">def list_sessions(self, limit=20):
     &#34;&#34;&#34;Prints a list of sessions using printer.table.&#34;&#34;&#34;
     sessions = self._get_sessions()
     if not sessions:
         printer.info(&#34;No saved AI sessions found.&#34;)
         return
     
+    total = len(sessions)
+    if limit and total &gt; limit:
+        sessions = sessions[:limit]
+        
     columns = [&#34;ID&#34;, &#34;Title&#34;, &#34;Created At&#34;, &#34;Model&#34;]
     rows = [[s[&#34;id&#34;], s[&#34;title&#34;], s[&#34;created_at&#34;], s[&#34;model&#34;]] for s in sessions]
-    printer.table(&#34;AI Persisted Sessions&#34;, columns, rows)</code></pre>
+    
+    title = &#34;AI Persisted Sessions&#34;
+    if limit and total &gt; limit:
+        title += f&#34; (Showing last {limit} of {total})&#34;
+        
+    printer.table(title, columns, rows)
+    if limit and total &gt; limit:
+        printer.info(f&#34;Use &#39;--list --all&#39; (if supported) or check the sessions directory to see all {total} sessions.&#34;)</code></pre>
 </details>
 <div class="desc"><p>Prints a list of sessions using printer.table.</p></div>
 </dd>
@@ -3070,6 +3141,8 @@ def confirm(self, user_input): return True</code></pre>
         first_user_msg = next((m[&#34;content&#34;] for m in history if m[&#34;role&#34;] == &#34;user&#34;), &#34;new-session&#34;)
         self.session_id = self._generate_session_id(first_user_msg)
         self.session_path = os.path.join(self.sessions_dir, f&#34;{self.session_id}.json&#34;)
+    elif not self.session_path:
+        self.session_path = os.path.join(self.sessions_dir, f&#34;{self.session_id}.json&#34;)
 
     # If it&#39;s a new file, we might want to set a better title
     if not os.path.exists(self.session_path) and not title:
@@ -4226,8 +4299,11 @@ class node:
 
 
     def _setup_interact_environment(self, debug=False, logger=None, async_mode=False):
-        size = re.search(&#39;columns=([0-9]+).*lines=([0-9]+)&#39;,str(os.get_terminal_size()))
+        try:
-        self.child.setwinsize(int(size.group(2)),int(size.group(1)))
+            size = re.search(&#39;columns=([0-9]+).*lines=([0-9]+)&#39;,str(os.get_terminal_size()))
+            self.child.setwinsize(int(size.group(2)),int(size.group(1)))
+        except OSError:
+            pass
         if logger:
             port_str = f&#34;:{self.port}&#34; if self.port and self.protocol not in [&#34;ssm&#34;, &#34;kubectl&#34;, &#34;docker&#34;] else &#34;&#34;
             logger(&#34;success&#34;, f&#34;Connected to {self.unique} at {self.host}{port_str} via: {self.protocol}&#34;)
@@ -4264,6 +4340,7 @@ class node:
 
     async def _async_interact_loop(self, local_stream, resize_callback, copilot_handler=None):
         local_stream.setup(resize_callback=resize_callback)
+        self.current_local_stream = local_stream
         try:
             child_fd = self.child.child_fd
             
@@ -4346,11 +4423,19 @@ class node:
                     # Remove any stray \x00 bytes and forward normally
                     clean_data = data.replace(b&#39;\x00&#39;, b&#39;&#39;)
                     if clean_data:
-                        # Track command boundaries when user hits Enter
+                        # Track command boundaries when user hits Enter or presses Ctrl+C
-                        if hasattr(self, &#39;mylog&#39;) and (b&#39;\r&#39; in clean_data or b&#39;\n&#39; in clean_data):
+                        if hasattr(self, &#39;mylog&#39;) and (b&#39;\r&#39; in clean_data or b&#39;\n&#39; in clean_data or b&#39;\x03&#39; in clean_data):
-                            self.cmd_byte_positions.append((self.mylog.tell(), None))
+                            pos = self.mylog.tell()
+                            marker_cmd = &#34;CANCELLED&#34; if b&#39;\x03&#39; in clean_data else None
+                            self.cmd_byte_positions.append((pos, marker_cmd))
+                            if hasattr(self, &#39;current_local_stream&#39;) and self.current_local_stream is not None:
+                                try:
+                                    await self.current_local_stream.write(f&#39;\x1b]133;B;{pos}\x07&#39;.encode())
+                                except Exception:
+                                    pass
 
-                        try:                            os.write(child_fd, clean_data)
+                        try:
+                            os.write(child_fd, clean_data)
                         except OSError:
                             break
                         self.lastinput = time()
@@ -4470,6 +4555,7 @@ class node:
                 except Exception:
                     pass
         finally:
+            self.current_local_stream = None
             local_stream.teardown()
 
     @MethodHook
@@ -4498,6 +4584,11 @@ class node:
             if cmd != slc and hasattr(self, &#39;cmd_byte_positions&#39;) and self.cmd_byte_positions is not None:
                 log_pos = self.mylog.tell() if hasattr(self, &#39;mylog&#39;) else 0
                 self.cmd_byte_positions.append((log_pos, cmd))
+                if hasattr(self, &#39;current_local_stream&#39;) and self.current_local_stream is not None:
+                    try:
+                        await self.current_local_stream.write(f&#39;\x1b]133;B;{log_pos}\x07&#39;.encode())
+                    except Exception:
+                        pass
             
             # Write physically to PTY
             os.write(child_fd, (cmd + &#34;\n&#34;).encode())
@@ -5137,6 +5228,11 @@ async def inject_commands(self, commands, child_fd, on_inject=None):
         if cmd != slc and hasattr(self, &#39;cmd_byte_positions&#39;) and self.cmd_byte_positions is not None:
             log_pos = self.mylog.tell() if hasattr(self, &#39;mylog&#39;) else 0
             self.cmd_byte_positions.append((log_pos, cmd))
+            if hasattr(self, &#39;current_local_stream&#39;) and self.current_local_stream is not None:
+                try:
+                    await self.current_local_stream.write(f&#39;\x1b]133;B;{log_pos}\x07&#39;.encode())
+                except Exception:
+                    pass
         
         # Write physically to PTY
         os.write(child_fd, (cmd + &#34;\n&#34;).encode())
@@ -6225,7 +6321,6 @@ def test(self, commands, expected, vars = None,*, folder = None, prompt = None,
 <li><code><a title="connpy.mcp_client" href="mcp_client.html">connpy.mcp_client</a></code></li>
 <li><code><a title="connpy.proto" href="proto/index.html">connpy.proto</a></code></li>
 <li><code><a title="connpy.services" href="services/index.html">connpy.services</a></code></li>
-<li><code><a title="connpy.tests" href="tests/index.html">connpy.tests</a></code></li>
 <li><code><a title="connpy.tunnels" href="tunnels.html">connpy.tunnels</a></code></li>
 <li><code><a title="connpy.utils" href="utils.html">connpy.utils</a></code></li>
 </ul>
@@ -6289,7 +6384,7 @@ def test(self, commands, expected, vars = None,*, folder = None, prompt = None,
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/mcp_client.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.mcp_client API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -343,7 +343,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/proto/index.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.proto API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -60,7 +60,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/ai_service.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services.ai_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -58,6 +58,37 @@ el.replaceWith(d);
 <pre><code class="python">class AIService(BaseService):
     &#34;&#34;&#34;Business logic for interacting with AI agents and LLM configurations.&#34;&#34;&#34;
 
+    def _clean_cisco_scrolling(self, text: str) -&gt; str:
+        &#34;&#34;&#34;Resolves horizontal scrolling artifacts (backspaces, \r, ANSI) by merging overlapping segments.&#34;&#34;&#34;
+        def merge_overlapping(s1, s2):
+            s2_clean = s2.lstrip(&#39; $&#39;)
+            max_overlap = min(len(s1), len(s2_clean))
+            for i in range(max_overlap, 0, -1):
+                if s1[-i:] == s2_clean[:i]:
+                    return s1 + s2_clean[i:]
+            return s1 + s2_clean
+
+        scroll_re = re.compile(r&#39;(\x08{5,}\s*\$?|\$\r|\x1b\[\d+[GD]\s*\$?)&#39;)
+        parts = scroll_re.split(text)
+        merged = &#34;&#34;
+        
+        for part in parts:
+            if scroll_re.match(part):
+                continue
+                
+            cleaned = log_cleaner(part)
+            if not merged:
+                merged = cleaned
+            else:
+                merged_lines = merged.split(&#39;\n&#39;)
+                cleaned_lines = cleaned.split(&#39;\n&#39;)
+                
+                merged_lines[-1] = merge_overlapping(merged_lines[-1], cleaned_lines[0])
+                merged_lines.extend(cleaned_lines[1:])
+                merged = &#34;\n&#34;.join(merged_lines)
+                
+        return merged
+
     def build_context_blocks(self, raw_bytes: bytes, cmd_byte_positions: list, node_info: dict, last_line: str = &#34;&#34;) -&gt; list:
         &#34;&#34;&#34;Identifies command blocks in the terminal history.&#34;&#34;&#34;
         blocks = []
@@ -79,28 +110,69 @@ el.replaceWith(d);
                 prev_pos = cmd_byte_positions[i-1][0]
                 
                 if known_cmd:
-                    prev_chunk = raw_bytes[prev_pos:pos]
+                    if known_cmd == &#34;CANCELLED&#34;:
-                    prev_cleaned = log_cleaner(prev_chunk.decode(errors=&#39;replace&#39;))
+                        parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;CANCELLED&#34;, &#34;preview&#34;: &#34;&#34;})
-                    prev_lines = [l for l in prev_cleaned.split(&#39;\n&#39;) if l.strip()]
+                    else:
-                    prompt_text = prev_lines[-1].strip() if prev_lines else &#34;&#34;
+                        prev_chunk = raw_bytes[prev_pos:pos]
-                    preview = f&#34;{prompt_text}{known_cmd}&#34; if prompt_text else known_cmd
+                        prev_cleaned = self._clean_cisco_scrolling(prev_chunk.decode(errors=&#39;replace&#39;))
-                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
+                        prev_lines = [l for l in prev_cleaned.split(&#39;\n&#39;) if l.strip()]
+                        prompt_text = prev_lines[-1].strip() if prev_lines else &#34;&#34;
+                        preview = f&#34;{prompt_text}{known_cmd}&#34; if prompt_text else known_cmd
+                        
+                        if len(preview) &gt; 80:
+                            preview = preview[:77] + &#34;...&#34;
+                        parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview})
                 else:
                     chunk = raw_bytes[prev_pos:pos]
-                    cleaned = log_cleaner(chunk.decode(errors=&#39;replace&#39;))
-                    lines = [l for l in cleaned.split(&#39;\n&#39;) if l.strip()]
-                    preview = lines[-1].strip() if lines else &#34;&#34;
                     
-                    if preview:
+                    cleaned = self._clean_cisco_scrolling(chunk.decode(errors=&#39;replace&#39;))
-                        match = prompt_re.search(preview)
+                    lines = [l for l in cleaned.split(&#39;\n&#39;) if l.strip()]
-                        if match:
+                    
-                            cmd_text = preview[match.end():].strip()
+                    found_in_pass1 = False
-                            if cmd_text:
+                    if lines:
-                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
+                        # Search backwards through the last few lines for the prompt
-                            else:
+                        for idx in range(len(lines) - 1, max(-1, len(lines) - 10), -1):
-                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;EMPTY_PROMPT&#34;, &#34;preview&#34;: &#34;&#34;})
+                            match = prompt_re.search(lines[idx])
-                        else:
+                            if match:
-                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
+                                ptxt = match.group(0).strip()
+                                cmd_first_line = lines[idx][match.end():].strip()
+                                cmd_rest = [l.strip() for l in lines[idx+1:]]
+                                cmd_text = &#34; &#34;.join([cmd_first_line] + cmd_rest).strip()
+                                
+                                if cmd_text:
+                                    pv = f&#34;{ptxt} {cmd_text}&#34;.strip()
+                                    if len(pv) &gt; 80:
+                                        pv = pv[:77] + &#34;...&#34;
+                                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: pv})
+                                else:
+                                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;EMPTY_PROMPT&#34;, &#34;preview&#34;: &#34;&#34;})
+                                found_in_pass1 = True
+                                break
+                        
+                        if not found_in_pass1:
+                            # Fallback: The prompt might have been isolated in the previous chunk 
+                            # due to asynchronous network delays splitting the output exactly at the newline.
+                            prev_was_valid_cmd = i &gt;= 2 and parsed_positions[i-2][&#34;type&#34;] == &#34;VALID_CMD&#34;
+                            if prev_pos &gt; 0 and not prev_was_valid_cmd:
+                                # Fetch the very last chunk that we just processed
+                                prev_prev_pos = cmd_byte_positions[i-2][0] if i &gt;= 2 else 0
+                                prev_chunk_text = self._clean_cisco_scrolling(raw_bytes[prev_prev_pos:prev_pos].decode(errors=&#39;replace&#39;))
+                                prev_lines_text = [l for l in prev_chunk_text.split(&#39;\n&#39;) if l.strip()]
+                                
+                                if prev_lines_text:
+                                    prev_match = prompt_re.search(prev_lines_text[-1])
+                                    if prev_match:
+                                        ptxt = prev_match.group(0).strip()
+                                        cmd_text = &#34; &#34;.join([l.strip() for l in lines]).strip()
+                                        if cmd_text:
+                                            pv = f&#34;{ptxt} {cmd_text}&#34;.strip()
+                                            if len(pv) &gt; 80:
+                                                pv = pv[:77] + &#34;...&#34;
+                                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: pv})
+                                            found_in_pass1 = True
+                            
+                            if not found_in_pass1:
+                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
                     else:
                         parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
 
@@ -113,11 +185,11 @@ el.replaceWith(d);
                 start_pos = item[&#34;pos&#34;]
                 preview = item[&#34;preview&#34;]
                 
-                # Find the end position: next VALID_CMD or EMPTY_PROMPT
+                # Find the end position: next VALID_CMD or EMPTY_PROMPT or CANCELLED
                 end_pos = current_prompt_pos
                 for j in range(i + 1, len(parsed_positions)):
                     next_item = parsed_positions[j]
-                    if next_item[&#34;type&#34;] in (&#34;VALID_CMD&#34;, &#34;EMPTY_PROMPT&#34;):
+                    if next_item[&#34;type&#34;] in (&#34;VALID_CMD&#34;, &#34;EMPTY_PROMPT&#34;, &#34;CANCELLED&#34;):
                         end_pos = next_item[&#34;pos&#34;]
                         break
                 
@@ -219,11 +291,14 @@ el.replaceWith(d);
         return await asyncio.wrap_future(future)
 
 
-    def list_sessions(self):
+    def list_sessions(self, limit=None):
-        &#34;&#34;&#34;Return a list of all saved AI sessions.&#34;&#34;&#34;
+        &#34;&#34;&#34;Return a list of saved AI sessions, optionally limited.&#34;&#34;&#34;
         from connpy.ai import ai
         agent = ai(self.config)
-        return agent._get_sessions()
+        sessions = agent._get_sessions()
+        if limit and len(sessions) &gt; limit:
+            return sessions[:limit], len(sessions)
+        return sessions, len(sessions)
 
     def delete_session(self, session_id):
         &#34;&#34;&#34;Delete an AI session by ID.&#34;&#34;&#34;
@@ -235,13 +310,15 @@ el.replaceWith(d);
         else:
             raise InvalidConfigurationError(f&#34;Session &#39;{session_id}&#39; not found.&#34;)
 
-    def configure_provider(self, provider, model=None, api_key=None):
+    def configure_provider(self, provider, model=None, api_key=None, auth=None):
         &#34;&#34;&#34;Update AI provider settings in the configuration.&#34;&#34;&#34;
         settings = self.config.config.get(&#34;ai&#34;, {})
         if model:
             settings[f&#34;{provider}_model&#34;] = model
         if api_key:
             settings[f&#34;{provider}_api_key&#34;] = api_key
+        if auth is not None:
+            settings[f&#34;{provider}_auth&#34;] = auth
             
         self.config.config[&#34;ai&#34;] = settings
         self.config._saveconfig(self.config.file)
@@ -280,6 +357,11 @@ el.replaceWith(d);
         self.config.config[&#34;ai&#34;] = ai_settings
         self.config._saveconfig(self.config.file)
 
+    def list_mcp_servers(self) -&gt; dict:
+        &#34;&#34;&#34;Get the configured MCP servers.&#34;&#34;&#34;
+        ai_settings = self.config.config.get(&#34;ai&#34;, {})
+        return ai_settings.get(&#34;mcp_servers&#34;, {})
+
     def load_session_data(self, session_id):
         &#34;&#34;&#34;Load a session&#39;s raw data by ID.&#34;&#34;&#34;
         from connpy.ai import ai
@@ -379,28 +461,69 @@ el.replaceWith(d);
             prev_pos = cmd_byte_positions[i-1][0]
             
             if known_cmd:
-                prev_chunk = raw_bytes[prev_pos:pos]
+                if known_cmd == &#34;CANCELLED&#34;:
-                prev_cleaned = log_cleaner(prev_chunk.decode(errors=&#39;replace&#39;))
+                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;CANCELLED&#34;, &#34;preview&#34;: &#34;&#34;})
-                prev_lines = [l for l in prev_cleaned.split(&#39;\n&#39;) if l.strip()]
+                else:
-                prompt_text = prev_lines[-1].strip() if prev_lines else &#34;&#34;
+                    prev_chunk = raw_bytes[prev_pos:pos]
-                preview = f&#34;{prompt_text}{known_cmd}&#34; if prompt_text else known_cmd
+                    prev_cleaned = self._clean_cisco_scrolling(prev_chunk.decode(errors=&#39;replace&#39;))
-                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
+                    prev_lines = [l for l in prev_cleaned.split(&#39;\n&#39;) if l.strip()]
+                    prompt_text = prev_lines[-1].strip() if prev_lines else &#34;&#34;
+                    preview = f&#34;{prompt_text}{known_cmd}&#34; if prompt_text else known_cmd
+                    
+                    if len(preview) &gt; 80:
+                        preview = preview[:77] + &#34;...&#34;
+                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview})
             else:
                 chunk = raw_bytes[prev_pos:pos]
-                cleaned = log_cleaner(chunk.decode(errors=&#39;replace&#39;))
-                lines = [l for l in cleaned.split(&#39;\n&#39;) if l.strip()]
-                preview = lines[-1].strip() if lines else &#34;&#34;
                 
-                if preview:
+                cleaned = self._clean_cisco_scrolling(chunk.decode(errors=&#39;replace&#39;))
-                    match = prompt_re.search(preview)
+                lines = [l for l in cleaned.split(&#39;\n&#39;) if l.strip()]
-                    if match:
+                
-                        cmd_text = preview[match.end():].strip()
+                found_in_pass1 = False
-                        if cmd_text:
+                if lines:
-                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
+                    # Search backwards through the last few lines for the prompt
-                        else:
+                    for idx in range(len(lines) - 1, max(-1, len(lines) - 10), -1):
-                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;EMPTY_PROMPT&#34;, &#34;preview&#34;: &#34;&#34;})
+                        match = prompt_re.search(lines[idx])
-                    else:
+                        if match:
-                        parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
+                            ptxt = match.group(0).strip()
+                            cmd_first_line = lines[idx][match.end():].strip()
+                            cmd_rest = [l.strip() for l in lines[idx+1:]]
+                            cmd_text = &#34; &#34;.join([cmd_first_line] + cmd_rest).strip()
+                            
+                            if cmd_text:
+                                pv = f&#34;{ptxt} {cmd_text}&#34;.strip()
+                                if len(pv) &gt; 80:
+                                    pv = pv[:77] + &#34;...&#34;
+                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: pv})
+                            else:
+                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;EMPTY_PROMPT&#34;, &#34;preview&#34;: &#34;&#34;})
+                            found_in_pass1 = True
+                            break
+                    
+                    if not found_in_pass1:
+                        # Fallback: The prompt might have been isolated in the previous chunk 
+                        # due to asynchronous network delays splitting the output exactly at the newline.
+                        prev_was_valid_cmd = i &gt;= 2 and parsed_positions[i-2][&#34;type&#34;] == &#34;VALID_CMD&#34;
+                        if prev_pos &gt; 0 and not prev_was_valid_cmd:
+                            # Fetch the very last chunk that we just processed
+                            prev_prev_pos = cmd_byte_positions[i-2][0] if i &gt;= 2 else 0
+                            prev_chunk_text = self._clean_cisco_scrolling(raw_bytes[prev_prev_pos:prev_pos].decode(errors=&#39;replace&#39;))
+                            prev_lines_text = [l for l in prev_chunk_text.split(&#39;\n&#39;) if l.strip()]
+                            
+                            if prev_lines_text:
+                                prev_match = prompt_re.search(prev_lines_text[-1])
+                                if prev_match:
+                                    ptxt = prev_match.group(0).strip()
+                                    cmd_text = &#34; &#34;.join([l.strip() for l in lines]).strip()
+                                    if cmd_text:
+                                        pv = f&#34;{ptxt} {cmd_text}&#34;.strip()
+                                        if len(pv) &gt; 80:
+                                            pv = pv[:77] + &#34;...&#34;
+                                        parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: pv})
+                                        found_in_pass1 = True
+                        
+                        if not found_in_pass1:
+                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
                 else:
                     parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
 
@@ -413,11 +536,11 @@ el.replaceWith(d);
             start_pos = item[&#34;pos&#34;]
             preview = item[&#34;preview&#34;]
             
-            # Find the end position: next VALID_CMD or EMPTY_PROMPT
+            # Find the end position: next VALID_CMD or EMPTY_PROMPT or CANCELLED
             end_pos = current_prompt_pos
             for j in range(i + 1, len(parsed_positions)):
                 next_item = parsed_positions[j]
-                if next_item[&#34;type&#34;] in (&#34;VALID_CMD&#34;, &#34;EMPTY_PROMPT&#34;):
+                if next_item[&#34;type&#34;] in (&#34;VALID_CMD&#34;, &#34;EMPTY_PROMPT&#34;, &#34;CANCELLED&#34;):
                     end_pos = next_item[&#34;pos&#34;]
                     break
             
@@ -478,20 +601,22 @@ el.replaceWith(d);
 <div class="desc"><p>Update MCP server settings in the configuration with smart merging.</p></div>
 </dd>
 <dt id="connpy.services.ai_service.AIService.configure_provider"><code class="name flex">
-<span>def <span class="ident">configure_provider</span></span>(<span>self, provider, model=None, api_key=None)</span>
+<span>def <span class="ident">configure_provider</span></span>(<span>self, provider, model=None, api_key=None, auth=None)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
-<pre><code class="python">def configure_provider(self, provider, model=None, api_key=None):
+<pre><code class="python">def configure_provider(self, provider, model=None, api_key=None, auth=None):
     &#34;&#34;&#34;Update AI provider settings in the configuration.&#34;&#34;&#34;
     settings = self.config.config.get(&#34;ai&#34;, {})
     if model:
         settings[f&#34;{provider}_model&#34;] = model
     if api_key:
         settings[f&#34;{provider}_api_key&#34;] = api_key
+    if auth is not None:
+        settings[f&#34;{provider}_auth&#34;] = auth
         
     self.config.config[&#34;ai&#34;] = settings
     self.config._saveconfig(self.config.file)</code></pre>
@@ -534,21 +659,39 @@ el.replaceWith(d);
 </details>
 <div class="desc"><p>Delete an AI session by ID.</p></div>
 </dd>
-<dt id="connpy.services.ai_service.AIService.list_sessions"><code class="name flex">
+<dt id="connpy.services.ai_service.AIService.list_mcp_servers"><code class="name flex">
-<span>def <span class="ident">list_sessions</span></span>(<span>self)</span>
+<span>def <span class="ident">list_mcp_servers</span></span>(<span>self) ‑> dict</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
-<pre><code class="python">def list_sessions(self):
+<pre><code class="python">def list_mcp_servers(self) -&gt; dict:
-    &#34;&#34;&#34;Return a list of all saved AI sessions.&#34;&#34;&#34;
+    &#34;&#34;&#34;Get the configured MCP servers.&#34;&#34;&#34;
+    ai_settings = self.config.config.get(&#34;ai&#34;, {})
+    return ai_settings.get(&#34;mcp_servers&#34;, {})</code></pre>
+</details>
+<div class="desc"><p>Get the configured MCP servers.</p></div>
+</dd>
+<dt id="connpy.services.ai_service.AIService.list_sessions"><code class="name flex">
+<span>def <span class="ident">list_sessions</span></span>(<span>self, limit=None)</span>
+</code></dt>
+<dd>
+<details class="source">
+<summary>
+<span>Expand source code</span>
+</summary>
+<pre><code class="python">def list_sessions(self, limit=None):
+    &#34;&#34;&#34;Return a list of saved AI sessions, optionally limited.&#34;&#34;&#34;
     from connpy.ai import ai
     agent = ai(self.config)
-    return agent._get_sessions()</code></pre>
+    sessions = agent._get_sessions()
+    if limit and len(sessions) &gt; limit:
+        return sessions[:limit], len(sessions)
+    return sessions, len(sessions)</code></pre>
 </details>
-<div class="desc"><p>Return a list of all saved AI sessions.</p></div>
+<div class="desc"><p>Return a list of saved AI sessions, optionally limited.</p></div>
 </dd>
 <dt id="connpy.services.ai_service.AIService.load_session_data"><code class="name flex">
 <span>def <span class="ident">load_session_data</span></span>(<span>self, session_id)</span>
@@ -671,6 +814,7 @@ el.replaceWith(d);
 <li><code><a title="connpy.services.ai_service.AIService.configure_provider" href="#connpy.services.ai_service.AIService.configure_provider">configure_provider</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.confirm" href="#connpy.services.ai_service.AIService.confirm">confirm</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.delete_session" href="#connpy.services.ai_service.AIService.delete_session">delete_session</a></code></li>
+<li><code><a title="connpy.services.ai_service.AIService.list_mcp_servers" href="#connpy.services.ai_service.AIService.list_mcp_servers">list_mcp_servers</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.list_sessions" href="#connpy.services.ai_service.AIService.list_sessions">list_sessions</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.load_session_data" href="#connpy.services.ai_service.AIService.load_session_data">load_session_data</a></code></li>
 <li><code><a title="connpy.services.ai_service.AIService.process_copilot_input" href="#connpy.services.ai_service.AIService.process_copilot_input">process_copilot_input</a></code></li>
@@ -682,7 +826,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/base.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services.base API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -152,7 +152,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/config_service.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services.config_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -319,7 +319,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/context_service.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services.context_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -370,7 +370,7 @@ def current_context(self) -&gt; str:
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/exceptions.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services.exceptions API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -268,7 +268,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/execution_service.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services.execution_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -449,7 +449,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/import_export_service.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services.import_export_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -361,7 +361,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/index.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -113,6 +113,37 @@ el.replaceWith(d);
 <pre><code class="python">class AIService(BaseService):
     &#34;&#34;&#34;Business logic for interacting with AI agents and LLM configurations.&#34;&#34;&#34;
 
+    def _clean_cisco_scrolling(self, text: str) -&gt; str:
+        &#34;&#34;&#34;Resolves horizontal scrolling artifacts (backspaces, \r, ANSI) by merging overlapping segments.&#34;&#34;&#34;
+        def merge_overlapping(s1, s2):
+            s2_clean = s2.lstrip(&#39; $&#39;)
+            max_overlap = min(len(s1), len(s2_clean))
+            for i in range(max_overlap, 0, -1):
+                if s1[-i:] == s2_clean[:i]:
+                    return s1 + s2_clean[i:]
+            return s1 + s2_clean
+
+        scroll_re = re.compile(r&#39;(\x08{5,}\s*\$?|\$\r|\x1b\[\d+[GD]\s*\$?)&#39;)
+        parts = scroll_re.split(text)
+        merged = &#34;&#34;
+        
+        for part in parts:
+            if scroll_re.match(part):
+                continue
+                
+            cleaned = log_cleaner(part)
+            if not merged:
+                merged = cleaned
+            else:
+                merged_lines = merged.split(&#39;\n&#39;)
+                cleaned_lines = cleaned.split(&#39;\n&#39;)
+                
+                merged_lines[-1] = merge_overlapping(merged_lines[-1], cleaned_lines[0])
+                merged_lines.extend(cleaned_lines[1:])
+                merged = &#34;\n&#34;.join(merged_lines)
+                
+        return merged
+
     def build_context_blocks(self, raw_bytes: bytes, cmd_byte_positions: list, node_info: dict, last_line: str = &#34;&#34;) -&gt; list:
         &#34;&#34;&#34;Identifies command blocks in the terminal history.&#34;&#34;&#34;
         blocks = []
@@ -134,28 +165,69 @@ el.replaceWith(d);
                 prev_pos = cmd_byte_positions[i-1][0]
                 
                 if known_cmd:
-                    prev_chunk = raw_bytes[prev_pos:pos]
+                    if known_cmd == &#34;CANCELLED&#34;:
-                    prev_cleaned = log_cleaner(prev_chunk.decode(errors=&#39;replace&#39;))
+                        parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;CANCELLED&#34;, &#34;preview&#34;: &#34;&#34;})
-                    prev_lines = [l for l in prev_cleaned.split(&#39;\n&#39;) if l.strip()]
+                    else:
-                    prompt_text = prev_lines[-1].strip() if prev_lines else &#34;&#34;
+                        prev_chunk = raw_bytes[prev_pos:pos]
-                    preview = f&#34;{prompt_text}{known_cmd}&#34; if prompt_text else known_cmd
+                        prev_cleaned = self._clean_cisco_scrolling(prev_chunk.decode(errors=&#39;replace&#39;))
-                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
+                        prev_lines = [l for l in prev_cleaned.split(&#39;\n&#39;) if l.strip()]
+                        prompt_text = prev_lines[-1].strip() if prev_lines else &#34;&#34;
+                        preview = f&#34;{prompt_text}{known_cmd}&#34; if prompt_text else known_cmd
+                        
+                        if len(preview) &gt; 80:
+                            preview = preview[:77] + &#34;...&#34;
+                        parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview})
                 else:
                     chunk = raw_bytes[prev_pos:pos]
-                    cleaned = log_cleaner(chunk.decode(errors=&#39;replace&#39;))
-                    lines = [l for l in cleaned.split(&#39;\n&#39;) if l.strip()]
-                    preview = lines[-1].strip() if lines else &#34;&#34;
                     
-                    if preview:
+                    cleaned = self._clean_cisco_scrolling(chunk.decode(errors=&#39;replace&#39;))
-                        match = prompt_re.search(preview)
+                    lines = [l for l in cleaned.split(&#39;\n&#39;) if l.strip()]
-                        if match:
+                    
-                            cmd_text = preview[match.end():].strip()
+                    found_in_pass1 = False
-                            if cmd_text:
+                    if lines:
-                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
+                        # Search backwards through the last few lines for the prompt
-                            else:
+                        for idx in range(len(lines) - 1, max(-1, len(lines) - 10), -1):
-                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;EMPTY_PROMPT&#34;, &#34;preview&#34;: &#34;&#34;})
+                            match = prompt_re.search(lines[idx])
-                        else:
+                            if match:
-                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
+                                ptxt = match.group(0).strip()
+                                cmd_first_line = lines[idx][match.end():].strip()
+                                cmd_rest = [l.strip() for l in lines[idx+1:]]
+                                cmd_text = &#34; &#34;.join([cmd_first_line] + cmd_rest).strip()
+                                
+                                if cmd_text:
+                                    pv = f&#34;{ptxt} {cmd_text}&#34;.strip()
+                                    if len(pv) &gt; 80:
+                                        pv = pv[:77] + &#34;...&#34;
+                                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: pv})
+                                else:
+                                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;EMPTY_PROMPT&#34;, &#34;preview&#34;: &#34;&#34;})
+                                found_in_pass1 = True
+                                break
+                        
+                        if not found_in_pass1:
+                            # Fallback: The prompt might have been isolated in the previous chunk 
+                            # due to asynchronous network delays splitting the output exactly at the newline.
+                            prev_was_valid_cmd = i &gt;= 2 and parsed_positions[i-2][&#34;type&#34;] == &#34;VALID_CMD&#34;
+                            if prev_pos &gt; 0 and not prev_was_valid_cmd:
+                                # Fetch the very last chunk that we just processed
+                                prev_prev_pos = cmd_byte_positions[i-2][0] if i &gt;= 2 else 0
+                                prev_chunk_text = self._clean_cisco_scrolling(raw_bytes[prev_prev_pos:prev_pos].decode(errors=&#39;replace&#39;))
+                                prev_lines_text = [l for l in prev_chunk_text.split(&#39;\n&#39;) if l.strip()]
+                                
+                                if prev_lines_text:
+                                    prev_match = prompt_re.search(prev_lines_text[-1])
+                                    if prev_match:
+                                        ptxt = prev_match.group(0).strip()
+                                        cmd_text = &#34; &#34;.join([l.strip() for l in lines]).strip()
+                                        if cmd_text:
+                                            pv = f&#34;{ptxt} {cmd_text}&#34;.strip()
+                                            if len(pv) &gt; 80:
+                                                pv = pv[:77] + &#34;...&#34;
+                                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: pv})
+                                            found_in_pass1 = True
+                            
+                            if not found_in_pass1:
+                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
                     else:
                         parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
 
@@ -168,11 +240,11 @@ el.replaceWith(d);
                 start_pos = item[&#34;pos&#34;]
                 preview = item[&#34;preview&#34;]
                 
-                # Find the end position: next VALID_CMD or EMPTY_PROMPT
+                # Find the end position: next VALID_CMD or EMPTY_PROMPT or CANCELLED
                 end_pos = current_prompt_pos
                 for j in range(i + 1, len(parsed_positions)):
                     next_item = parsed_positions[j]
-                    if next_item[&#34;type&#34;] in (&#34;VALID_CMD&#34;, &#34;EMPTY_PROMPT&#34;):
+                    if next_item[&#34;type&#34;] in (&#34;VALID_CMD&#34;, &#34;EMPTY_PROMPT&#34;, &#34;CANCELLED&#34;):
                         end_pos = next_item[&#34;pos&#34;]
                         break
                 
@@ -274,11 +346,14 @@ el.replaceWith(d);
         return await asyncio.wrap_future(future)
 
 
-    def list_sessions(self):
+    def list_sessions(self, limit=None):
-        &#34;&#34;&#34;Return a list of all saved AI sessions.&#34;&#34;&#34;
+        &#34;&#34;&#34;Return a list of saved AI sessions, optionally limited.&#34;&#34;&#34;
         from connpy.ai import ai
         agent = ai(self.config)
-        return agent._get_sessions()
+        sessions = agent._get_sessions()
+        if limit and len(sessions) &gt; limit:
+            return sessions[:limit], len(sessions)
+        return sessions, len(sessions)
 
     def delete_session(self, session_id):
         &#34;&#34;&#34;Delete an AI session by ID.&#34;&#34;&#34;
@@ -290,13 +365,15 @@ el.replaceWith(d);
         else:
             raise InvalidConfigurationError(f&#34;Session &#39;{session_id}&#39; not found.&#34;)
 
-    def configure_provider(self, provider, model=None, api_key=None):
+    def configure_provider(self, provider, model=None, api_key=None, auth=None):
         &#34;&#34;&#34;Update AI provider settings in the configuration.&#34;&#34;&#34;
         settings = self.config.config.get(&#34;ai&#34;, {})
         if model:
             settings[f&#34;{provider}_model&#34;] = model
         if api_key:
             settings[f&#34;{provider}_api_key&#34;] = api_key
+        if auth is not None:
+            settings[f&#34;{provider}_auth&#34;] = auth
             
         self.config.config[&#34;ai&#34;] = settings
         self.config._saveconfig(self.config.file)
@@ -335,6 +412,11 @@ el.replaceWith(d);
         self.config.config[&#34;ai&#34;] = ai_settings
         self.config._saveconfig(self.config.file)
 
+    def list_mcp_servers(self) -&gt; dict:
+        &#34;&#34;&#34;Get the configured MCP servers.&#34;&#34;&#34;
+        ai_settings = self.config.config.get(&#34;ai&#34;, {})
+        return ai_settings.get(&#34;mcp_servers&#34;, {})
+
     def load_session_data(self, session_id):
         &#34;&#34;&#34;Load a session&#39;s raw data by ID.&#34;&#34;&#34;
         from connpy.ai import ai
@@ -434,28 +516,69 @@ el.replaceWith(d);
             prev_pos = cmd_byte_positions[i-1][0]
             
             if known_cmd:
-                prev_chunk = raw_bytes[prev_pos:pos]
+                if known_cmd == &#34;CANCELLED&#34;:
-                prev_cleaned = log_cleaner(prev_chunk.decode(errors=&#39;replace&#39;))
+                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;CANCELLED&#34;, &#34;preview&#34;: &#34;&#34;})
-                prev_lines = [l for l in prev_cleaned.split(&#39;\n&#39;) if l.strip()]
+                else:
-                prompt_text = prev_lines[-1].strip() if prev_lines else &#34;&#34;
+                    prev_chunk = raw_bytes[prev_pos:pos]
-                preview = f&#34;{prompt_text}{known_cmd}&#34; if prompt_text else known_cmd
+                    prev_cleaned = self._clean_cisco_scrolling(prev_chunk.decode(errors=&#39;replace&#39;))
-                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
+                    prev_lines = [l for l in prev_cleaned.split(&#39;\n&#39;) if l.strip()]
+                    prompt_text = prev_lines[-1].strip() if prev_lines else &#34;&#34;
+                    preview = f&#34;{prompt_text}{known_cmd}&#34; if prompt_text else known_cmd
+                    
+                    if len(preview) &gt; 80:
+                        preview = preview[:77] + &#34;...&#34;
+                    parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview})
             else:
                 chunk = raw_bytes[prev_pos:pos]
-                cleaned = log_cleaner(chunk.decode(errors=&#39;replace&#39;))
-                lines = [l for l in cleaned.split(&#39;\n&#39;) if l.strip()]
-                preview = lines[-1].strip() if lines else &#34;&#34;
                 
-                if preview:
+                cleaned = self._clean_cisco_scrolling(chunk.decode(errors=&#39;replace&#39;))
-                    match = prompt_re.search(preview)
+                lines = [l for l in cleaned.split(&#39;\n&#39;) if l.strip()]
-                    if match:
+                
-                        cmd_text = preview[match.end():].strip()
+                found_in_pass1 = False
-                        if cmd_text:
+                if lines:
-                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: preview[:80]})
+                    # Search backwards through the last few lines for the prompt
-                        else:
+                    for idx in range(len(lines) - 1, max(-1, len(lines) - 10), -1):
-                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;EMPTY_PROMPT&#34;, &#34;preview&#34;: &#34;&#34;})
+                        match = prompt_re.search(lines[idx])
-                    else:
+                        if match:
-                        parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
+                            ptxt = match.group(0).strip()
+                            cmd_first_line = lines[idx][match.end():].strip()
+                            cmd_rest = [l.strip() for l in lines[idx+1:]]
+                            cmd_text = &#34; &#34;.join([cmd_first_line] + cmd_rest).strip()
+                            
+                            if cmd_text:
+                                pv = f&#34;{ptxt} {cmd_text}&#34;.strip()
+                                if len(pv) &gt; 80:
+                                    pv = pv[:77] + &#34;...&#34;
+                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: pv})
+                            else:
+                                parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;EMPTY_PROMPT&#34;, &#34;preview&#34;: &#34;&#34;})
+                            found_in_pass1 = True
+                            break
+                    
+                    if not found_in_pass1:
+                        # Fallback: The prompt might have been isolated in the previous chunk 
+                        # due to asynchronous network delays splitting the output exactly at the newline.
+                        prev_was_valid_cmd = i &gt;= 2 and parsed_positions[i-2][&#34;type&#34;] == &#34;VALID_CMD&#34;
+                        if prev_pos &gt; 0 and not prev_was_valid_cmd:
+                            # Fetch the very last chunk that we just processed
+                            prev_prev_pos = cmd_byte_positions[i-2][0] if i &gt;= 2 else 0
+                            prev_chunk_text = self._clean_cisco_scrolling(raw_bytes[prev_prev_pos:prev_pos].decode(errors=&#39;replace&#39;))
+                            prev_lines_text = [l for l in prev_chunk_text.split(&#39;\n&#39;) if l.strip()]
+                            
+                            if prev_lines_text:
+                                prev_match = prompt_re.search(prev_lines_text[-1])
+                                if prev_match:
+                                    ptxt = prev_match.group(0).strip()
+                                    cmd_text = &#34; &#34;.join([l.strip() for l in lines]).strip()
+                                    if cmd_text:
+                                        pv = f&#34;{ptxt} {cmd_text}&#34;.strip()
+                                        if len(pv) &gt; 80:
+                                            pv = pv[:77] + &#34;...&#34;
+                                        parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;VALID_CMD&#34;, &#34;preview&#34;: pv})
+                                        found_in_pass1 = True
+                        
+                        if not found_in_pass1:
+                            parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
                 else:
                     parsed_positions.append({&#34;pos&#34;: pos, &#34;type&#34;: &#34;SCROLLING&#34;, &#34;preview&#34;: &#34;&#34;})
 
@@ -468,11 +591,11 @@ el.replaceWith(d);
             start_pos = item[&#34;pos&#34;]
             preview = item[&#34;preview&#34;]
             
-            # Find the end position: next VALID_CMD or EMPTY_PROMPT
+            # Find the end position: next VALID_CMD or EMPTY_PROMPT or CANCELLED
             end_pos = current_prompt_pos
             for j in range(i + 1, len(parsed_positions)):
                 next_item = parsed_positions[j]
-                if next_item[&#34;type&#34;] in (&#34;VALID_CMD&#34;, &#34;EMPTY_PROMPT&#34;):
+                if next_item[&#34;type&#34;] in (&#34;VALID_CMD&#34;, &#34;EMPTY_PROMPT&#34;, &#34;CANCELLED&#34;):
                     end_pos = next_item[&#34;pos&#34;]
                     break
             
@@ -533,20 +656,22 @@ el.replaceWith(d);
 <div class="desc"><p>Update MCP server settings in the configuration with smart merging.</p></div>
 </dd>
 <dt id="connpy.services.AIService.configure_provider"><code class="name flex">
-<span>def <span class="ident">configure_provider</span></span>(<span>self, provider, model=None, api_key=None)</span>
+<span>def <span class="ident">configure_provider</span></span>(<span>self, provider, model=None, api_key=None, auth=None)</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
-<pre><code class="python">def configure_provider(self, provider, model=None, api_key=None):
+<pre><code class="python">def configure_provider(self, provider, model=None, api_key=None, auth=None):
     &#34;&#34;&#34;Update AI provider settings in the configuration.&#34;&#34;&#34;
     settings = self.config.config.get(&#34;ai&#34;, {})
     if model:
         settings[f&#34;{provider}_model&#34;] = model
     if api_key:
         settings[f&#34;{provider}_api_key&#34;] = api_key
+    if auth is not None:
+        settings[f&#34;{provider}_auth&#34;] = auth
         
     self.config.config[&#34;ai&#34;] = settings
     self.config._saveconfig(self.config.file)</code></pre>
@@ -589,21 +714,39 @@ el.replaceWith(d);
 </details>
 <div class="desc"><p>Delete an AI session by ID.</p></div>
 </dd>
-<dt id="connpy.services.AIService.list_sessions"><code class="name flex">
+<dt id="connpy.services.AIService.list_mcp_servers"><code class="name flex">
-<span>def <span class="ident">list_sessions</span></span>(<span>self)</span>
+<span>def <span class="ident">list_mcp_servers</span></span>(<span>self) ‑> dict</span>
 </code></dt>
 <dd>
 <details class="source">
 <summary>
 <span>Expand source code</span>
 </summary>
-<pre><code class="python">def list_sessions(self):
+<pre><code class="python">def list_mcp_servers(self) -&gt; dict:
-    &#34;&#34;&#34;Return a list of all saved AI sessions.&#34;&#34;&#34;
+    &#34;&#34;&#34;Get the configured MCP servers.&#34;&#34;&#34;
+    ai_settings = self.config.config.get(&#34;ai&#34;, {})
+    return ai_settings.get(&#34;mcp_servers&#34;, {})</code></pre>
+</details>
+<div class="desc"><p>Get the configured MCP servers.</p></div>
+</dd>
+<dt id="connpy.services.AIService.list_sessions"><code class="name flex">
+<span>def <span class="ident">list_sessions</span></span>(<span>self, limit=None)</span>
+</code></dt>
+<dd>
+<details class="source">
+<summary>
+<span>Expand source code</span>
+</summary>
+<pre><code class="python">def list_sessions(self, limit=None):
+    &#34;&#34;&#34;Return a list of saved AI sessions, optionally limited.&#34;&#34;&#34;
     from connpy.ai import ai
     agent = ai(self.config)
-    return agent._get_sessions()</code></pre>
+    sessions = agent._get_sessions()
+    if limit and len(sessions) &gt; limit:
+        return sessions[:limit], len(sessions)
+    return sessions, len(sessions)</code></pre>
 </details>
-<div class="desc"><p>Return a list of all saved AI sessions.</p></div>
+<div class="desc"><p>Return a list of saved AI sessions, optionally limited.</p></div>
 </dd>
 <dt id="connpy.services.AIService.load_session_data"><code class="name flex">
 <span>def <span class="ident">load_session_data</span></span>(<span>self, session_id)</span>
@@ -3726,6 +3869,7 @@ el.replaceWith(d);
 <li><code><a title="connpy.services.AIService.configure_provider" href="#connpy.services.AIService.configure_provider">configure_provider</a></code></li>
 <li><code><a title="connpy.services.AIService.confirm" href="#connpy.services.AIService.confirm">confirm</a></code></li>
 <li><code><a title="connpy.services.AIService.delete_session" href="#connpy.services.AIService.delete_session">delete_session</a></code></li>
+<li><code><a title="connpy.services.AIService.list_mcp_servers" href="#connpy.services.AIService.list_mcp_servers">list_mcp_servers</a></code></li>
 <li><code><a title="connpy.services.AIService.list_sessions" href="#connpy.services.AIService.list_sessions">list_sessions</a></code></li>
 <li><code><a title="connpy.services.AIService.load_session_data" href="#connpy.services.AIService.load_session_data">load_session_data</a></code></li>
 <li><code><a title="connpy.services.AIService.process_copilot_input" href="#connpy.services.AIService.process_copilot_input">process_copilot_input</a></code></li>
@@ -3840,7 +3984,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/node_service.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services.node_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -786,7 +786,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/plugin_service.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services.plugin_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -709,7 +709,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/profile_service.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services.profile_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -429,7 +429,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/provider.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services.provider API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -164,7 +164,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/sync_service.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services.sync_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -964,7 +964,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/services/system_service.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.services.system_service API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -325,7 +325,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/tunnels.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.tunnels API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -545,7 +545,7 @@ Bridges the blocking gRPC iterators with the async _async_interact_loop.</p></di
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>

docs/connpy/utils.html

@@ -3,7 +3,7 @@
 <head>
 <meta charset="utf-8">
 <meta name="viewport" content="width=device-width, initial-scale=1, minimum-scale=1">
-<meta name="generator" content="pdoc3 0.11.5">
+<meta name="generator" content="pdoc3 0.11.6">
 <title>connpy.utils API documentation</title>
 <meta name="description" content="">
 <link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/10up-sanitize.css/13.0.0/sanitize.min.css" integrity="sha512-y1dtMcuvtTMJc1yPgEqF0ZjQbhnc/bFhyvIyVNb9Zk5mIGtqVaAB1Ttl28su8AvFMOY0EwRbAe+HCLqj6W7/KA==" crossorigin>
@@ -59,11 +59,14 @@ el.replaceWith(d);
     if not data:
         return &#34;&#34;
             
+    # Remove OSC (Operating System Command) sequences (e.g., set window title \x1b]0;...\x07)
+    data = re.sub(r&#39;\x1b\][^\x07\x1b]*(?:\x07|\x1b\\)&#39;, &#39;&#39;, data)
+
     lines = data.split(&#39;\n&#39;)
     cleaned_lines = []
     
     # Regex to capture: ANSI sequences, control characters (\r, \b, etc), and plain text chunks
-    token_re = re.compile(r&#39;(\x1B(?:[@-Z\\-_]|\[[0-?]*[ -/ ]*[@-~])|\r|\b|\x7f|[\x00-\x1F]|[^\x1B\r\b\x7f\x00-\x1F]+)&#39;)
+    token_re = re.compile(r&#39;(\x1B(?:[\x30-\x5A\x5C-\x7E]|\[[0-?]*[ -/ ]*[@-~])|\r|\b|\x7f|[\x00-\x1F]|[^\x1B\r\b\x7f\x00-\x1F]+)&#39;)
     
     for line in lines:
         buffer = []
@@ -144,7 +147,7 @@ el.replaceWith(d);
 </nav>
 </main>
 <footer id="footer">
-<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.5</a>.</p>
+<p>Generated by <a href="https://pdoc3.github.io/pdoc" title="pdoc: Python API documentation generator"><cite>pdoc</cite> 0.11.6</a>.</p>
 </footer>
 </body>
 </html>