diff --git a/src/cai/repl/commands/completer.py b/src/cai/repl/commands/completer.py
index cadf97a5..3f6d2468 100644
--- a/src/cai/repl/commands/completer.py
+++ b/src/cai/repl/commands/completer.py
@@ -120,6 +120,9 @@ class FuzzyCommandCompleter(Completer):
 
             # Standard models always available
             standard_models = [
+                # Alias models
+                "alias0",
+
                 # Claude 3.7 models
                 "claude-3-7-sonnet-20250219",
 
diff --git a/src/cai/repl/commands/model.py b/src/cai/repl/commands/model.py
index 07cb0480..337fd9e9 100644
--- a/src/cai/repl/commands/model.py
+++ b/src/cai/repl/commands/model.py
@@ -67,11 +67,19 @@ class ModelCommand(Command):
         # Define model categories and their models for easy reference
         # pylint: disable=invalid-name
         MODEL_CATEGORIES = {
+            "Alias": [
+                {
+                    "name": "alias0",
+                    "description": (
+                        "Best model for Cybersecurity AI tasks"
+                    )
+                }
+            ],
             "Claude 3.7": [
                 {
                     "name": "claude-3-7-sonnet-20250219",
                     "description": (
-                        "Best model for complex reasoning and creative tasks"
+                        "Excellent model for complex reasoning and creative tasks"
                     )
                 }
             ],
diff --git a/src/cai/sdk/agents/models/openai_chatcompletions.py b/src/cai/sdk/agents/models/openai_chatcompletions.py
index 04de9fa8..9fb5b431 100644
--- a/src/cai/sdk/agents/models/openai_chatcompletions.py
+++ b/src/cai/sdk/agents/models/openai_chatcompletions.py
@@ -577,6 +577,10 @@ class OpenAIChatCompletionsModel(Model):
                 response.usage.input_tokens = response.usage.prompt_tokens
             if hasattr(response.usage, 'completion_tokens') and not hasattr(response.usage, 'output_tokens'):
                 response.usage.output_tokens = response.usage.completion_tokens
+                
+            # Ensure cost is properly initialized
+            if not hasattr(response, 'cost'):
+                response.cost = None
 
             return ModelResponse(
                 output=items,
diff --git a/src/cai/sdk/agents/run_to_jsonl.py b/src/cai/sdk/agents/run_to_jsonl.py
index 9a2a0e93..a8197090 100644
--- a/src/cai/sdk/agents/run_to_jsonl.py
+++ b/src/cai/sdk/agents/run_to_jsonl.py
@@ -158,7 +158,7 @@ class DataRecorder:  # pylint: disable=too-few-public-methods
         # Obtener el coste de la interacción
         interaction_cost = 0.0
         if hasattr(msg, "cost"):
-            interaction_cost = float(msg.cost)
+            interaction_cost = float(msg.cost) if msg.cost is not None else 0.0
 
         # Usar el total_cost proporcionado o actualizar el interno
         if total_cost is not None: