diff --git a/pilot/helpers/AgentConvo.py b/pilot/helpers/AgentConvo.py
index b8c5081..b5bac8e 100644
--- a/pilot/helpers/AgentConvo.py
+++ b/pilot/helpers/AgentConvo.py
@@ -124,7 +124,9 @@ class AgentConvo:
     def log_message(self, content):
         print_msg = capitalize_first_word_with_underscores(self.high_level_step)
         if self.log_to_user:
-            print(colored("Dev step ", 'yellow') + colored(self.agent.project.checkpoints['last_development_step'], 'yellow', attrs=['bold']) + f"\n{content}\n")
+            if self.agent.project.checkpoints['last_development_step'] is not None:
+                print(colored("\nDev step ", 'yellow') + colored(self.agent.project.checkpoints['last_development_step'], 'yellow', attrs=['bold']) + '\n', end='')
+            print(f"\n{content}\n")
         logger.info(f"{print_msg}: {content}\n")
 
     def to_playground(self):
diff --git a/pilot/helpers/Project.py b/pilot/helpers/Project.py
index 6eddb5e..2a4283c 100644
--- a/pilot/helpers/Project.py
+++ b/pilot/helpers/Project.py
@@ -181,14 +181,14 @@ class Project:
         delete_unconnected_steps_from(self.checkpoints['last_user_input'], 'previous_step')
 
     def ask_for_human_intervention(self, message, description=None, cbs={}):
-        print(colored(message, "yellow"))
+        print(colored(message, "yellow", attrs=['bold']))
         if description is not None:
             print(description)
         answer = ''
         while answer != 'continue':
             answer = styled_text(
                 self,
-                'Once you are ready, type "continue" to continue.',
+                'If something is wrong, tell me or type "continue" to continue.',
             )
 
             if answer in cbs:
diff --git a/pilot/helpers/agents/Developer.py b/pilot/helpers/agents/Developer.py
index 06ee8ba..cb451e4 100644
--- a/pilot/helpers/agents/Developer.py
+++ b/pilot/helpers/agents/Developer.py
@@ -26,7 +26,7 @@ class Developer(Agent):
             self.project.skip_steps = False if ('skip_until_dev_step' in self.project.args and self.project.args['skip_until_dev_step'] == '0') else True
 
         # DEVELOPMENT
-        print(colored(f"Ok, great, now, let's start with the actual development...\n", "green"))
+        print(colored(f"Ok, great, now, let's start with the actual development...\n", "green", attrs=['bold']))
         logger.info(f"Starting to create the actual code...")
 
         self.implement_task()
@@ -88,7 +88,7 @@ class Developer(Agent):
                 # TODO end
 
             elif step['type'] == 'human_intervention':
-                user_feedback = self.project.ask_for_human_intervention('I need your help! Can you try debugging this yourself and let me take over afterwards? Here are the details about the issue:', step['human_intervention_description'])
+                user_feedback = self.project.ask_for_human_intervention('I need human intervention:', step['human_intervention_description'])
                 if user_feedback is not None and user_feedback != 'continue':
                     debug(convo, user_input=user_feedback, issue_description=step['human_intervention_description'])
 
diff --git a/pilot/helpers/agents/ProductOwner.py b/pilot/helpers/agents/ProductOwner.py
index c6acceb..d592fb3 100644
--- a/pilot/helpers/agents/ProductOwner.py
+++ b/pilot/helpers/agents/ProductOwner.py
@@ -43,6 +43,7 @@ class ProductOwner(Agent):
             self.project,
             generate_messages_from_description(main_prompt, self.project.args['app_type'], self.project.args['name']))
 
+        print(colored('Project Summary:\n', 'green', attrs=['bold']))
         high_level_summary = convo_project_description.send_message('utils/summary.prompt',
             {'conversation': '\n'.join([f"{msg['role']}: {msg['content']}" for msg in high_level_messages])})
 
@@ -71,7 +72,7 @@ class ProductOwner(Agent):
             return step['user_stories']
 
         # USER STORIES
-        msg = f"Generating USER STORIES...\n"
+        msg = f"User Stories:\n"
         print(colored(msg, "green", attrs=['bold']))
         logger.info(msg)
 
@@ -105,7 +106,7 @@ class ProductOwner(Agent):
             return step['user_tasks']
 
         # USER TASKS
-        msg = f"Generating USER TASKS...\n"
+        msg = f"User Tasks:\n"
         print(colored(msg, "green", attrs=['bold']))
         logger.info(msg)
 
diff --git a/pilot/helpers/cli.py b/pilot/helpers/cli.py
index c34d8a1..c250a78 100644
--- a/pilot/helpers/cli.py
+++ b/pilot/helpers/cli.py
@@ -49,7 +49,8 @@ def execute_command(project, command, timeout=None, force=False):
         timeout = min(max(timeout, MIN_COMMAND_RUN_TIME), MAX_COMMAND_RUN_TIME)
 
     if not force:
-        print(colored(f'Can i execute the command: `') + colored(command, 'white', attrs=['bold']) + colored(f'` with {timeout}ms timeout?'))
+        print(colored(f'\n--------- EXECUTE COMMAND ----------', 'yellow', attrs=['bold']))
+        print(colored(f'Can i execute the command: `') + colored(command, 'yellow', attrs=['bold']) + colored(f'` with {timeout}ms timeout?'))
 
         answer = styled_text(
             project,
diff --git a/pilot/prompts/prompts.py b/pilot/prompts/prompts.py
index f2fe8cd..c667268 100644
--- a/pilot/prompts/prompts.py
+++ b/pilot/prompts/prompts.py
@@ -75,7 +75,7 @@ def get_additional_info_from_openai(project, messages):
 
         if response is not None:
             if response['text'].strip() == END_RESPONSE:
-                print(response['text'])
+                print(response['text'] + '\n')
                 return messages
 
             # Ask the question to the user
@@ -109,9 +109,7 @@ def get_additional_info_from_user(project,  messages, role):
             if answer.lower() == '':
                 break
             response = create_gpt_chat_completion(
-                generate_messages_from_custom_conversation(role, [get_prompt('utils/update.prompt'), message, answer],
-                                                           'user'),
-                'additional_info')
+                generate_messages_from_custom_conversation(role, [get_prompt('utils/update.prompt'), message, answer], 'user'), 'additional_info')
 
             message = response
 
diff --git a/pilot/utils/arguments.py b/pilot/utils/arguments.py
index 2546aca..8590373 100644
--- a/pilot/utils/arguments.py
+++ b/pilot/utils/arguments.py
@@ -49,6 +49,8 @@ def get_arguments():
     if 'step' not in arguments:
         arguments['step'] = None
 
+    print(colored('\n------------------ STARTING NEW PROJECT ----------------------', 'green', attrs=['bold']))
     print(f"If you wish to continue with this project in future run:")
-    print(colored(f'python main.py app_id={arguments["app_id"]}', 'yellow', attrs=['bold']))
+    print(colored(f'python main.py app_id={arguments["app_id"]}', 'green', attrs=['bold']))
+    print(colored('--------------------------------------------------------------\n', 'green', attrs=['bold']))
     return arguments
diff --git a/pilot/utils/llm_connection.py b/pilot/utils/llm_connection.py
index c35b2c0..c191d8e 100644
--- a/pilot/utils/llm_connection.py
+++ b/pilot/utils/llm_connection.py
@@ -74,8 +74,8 @@ def num_tokens_from_functions(functions, model="gpt-4"):
                             for o in v['enum']:
                                 function_tokens += 3
                                 function_tokens += len(encoding.encode(o))
-                        else:
-                            print(f"Warning: not supported field {field}")
+                        # else:
+                        #     print(f"Warning: not supported field {field}")
                 function_tokens += 11
 
         num_tokens += function_tokens
@@ -170,7 +170,7 @@ def stream_gpt_completion(data, req_type):
         return result_data
 
     # spinner = spinner_start(colored("Waiting for OpenAI API response...", 'yellow'))
-    print(colored("Waiting for OpenAI API response...", 'yellow'))
+    # print(colored("Stream response from OpenAI:", 'yellow'))
     api_key = os.getenv("OPENAI_API_KEY")
 
     logger.info(f'Request data: {data}')
diff --git a/pilot/utils/questionary.py b/pilot/utils/questionary.py
index d81f279..a9a9506 100644
--- a/pilot/utils/questionary.py
+++ b/pilot/utils/questionary.py
@@ -33,4 +33,7 @@ def styled_text(project, question):
     }
     response = questionary.text(question, **config).unsafe_ask()  # .ask() is included here
     user_input = save_user_input(project, question, response)
+
+    print('\n\n', end='')
+
     return response