diff --git a/autogpt/agent/agent.py b/autogpt/agent/agent.py
index 48d1932806a8..a7ea83239737 100644
--- a/autogpt/agent/agent.py
+++ b/autogpt/agent/agent.py
@@ -161,7 +161,7 @@ def start_interaction_loop(self):
                 )
 
                 logger.info(
-                    "Enter 'y' to authorise command, 'y -N' to run N continuous commands, 's' to run self-feedback commands"
+                    "Enter 'y' to authorise command, 'y -N' to run N continuous commands, 's' to run self-feedback commands or "
                     "'n' to exit program, or enter feedback for "
                     f"{self.ai_name}..."
                 )
@@ -190,10 +190,8 @@ def start_interaction_loop(self):
                             Fore.YELLOW,
                             "",
                         )
-                        if self_feedback_resp[0].lower().strip() == cfg.authorise_key:
-                            user_input = "GENERATE NEXT COMMAND JSON"
-                        else:
-                            user_input = self_feedback_resp
+                        user_input = self_feedback_resp
+                        command_name = "self_feedback"
                         break
                     elif console_input.lower().strip() == "":
                         logger.warn("Invalid input format.")
@@ -244,6 +242,8 @@ def start_interaction_loop(self):
                 )
             elif command_name == "human_feedback":
                 result = f"Human feedback: {user_input}"
+            elif command_name == "self_feedback":
+                result = f"Self feedback: {user_input}"
             else:
                 for plugin in cfg.plugins:
                     if not plugin.can_handle_pre_command():
@@ -314,12 +314,11 @@ def get_self_feedback(self, thoughts: dict, llm_model: str) -> str:
         """
         ai_role = self.config.ai_role
 
-        feedback_prompt = f"Below is a message from an AI agent with the role of {ai_role}. Please review the provided Thought, Reasoning, Plan, and Criticism. If these elements accurately contribute to the successful execution of the assumed role, respond with the letter 'Y' followed by a space, and then explain why it is effective. If the provided information is not suitable for achieving the role's objectives, please provide one or more sentences addressing the issue and suggesting a resolution."
+        feedback_prompt = f"Below is a message from me, an AI Agent, assuming the role of {ai_role}. whilst keeping knowledge of my slight limitations as an AI Agent Please evaluate my thought process, reasoning, and plan, and provide a concise paragraph outlining potential improvements. Consider adding or removing ideas that do not align with my role and explaining why, prioritizing thoughts based on their significance, or simply refining my overall thought process."
         reasoning = thoughts.get("reasoning", "")
         plan = thoughts.get("plan", "")
         thought = thoughts.get("thoughts", "")
-        criticism = thoughts.get("criticism", "")
-        feedback_thoughts = thought + reasoning + plan + criticism
+        feedback_thoughts = thought + reasoning + plan
         return create_chat_completion(
             [{"role": "user", "content": feedback_prompt + feedback_thoughts}],
             llm_model,
diff --git a/tests/unit/test_get_self_feedback.py b/tests/unit/test_get_self_feedback.py
index 09707270d34f..e1e9bd4ad4ee 100644
--- a/tests/unit/test_get_self_feedback.py
+++ b/tests/unit/test_get_self_feedback.py
@@ -9,12 +9,14 @@ def test_get_self_feedback(mocker):
         "reasoning": "Sample reasoning.",
         "plan": "Sample plan.",
         "thoughts": "Sample thoughts.",
-        "criticism": "Sample criticism.",
     }
 
     # Define a fake response for the create_chat_completion function
     fake_response = (
-        "Y The provided information is suitable for achieving the role's objectives."
+        "The AI Agent has demonstrated a reasonable thought process, but there is room for improvement. "
+        "For example, the reasoning could be elaborated to better justify the plan, and the plan itself "
+        "could be more detailed to ensure its effectiveness. In addition, the AI Agent should focus more "
+        "on its core role and prioritize thoughts that align with that role."
     )
 
     # Mock the create_chat_completion function
@@ -36,5 +38,9 @@ def test_get_self_feedback(mocker):
         "gpt-3.5-turbo",
     )
 
-    # Check if the response is correct
-    assert feedback == fake_response
+    # Check if the response is a non-empty string
+    assert isinstance(feedback, str) and len(feedback) > 0
+
+    # Check if certain keywords from input thoughts are present in the feedback response
+    for keyword in ["reasoning", "plan", "thoughts"]:
+        assert keyword in feedback