ding shaky tutor flow

agent.py

@@ -131,9 +131,22 @@ class QMLAgent():
 
 It is able to process and understand large amounts of text, and can use this knowledge to provide accurate and informative responses to a wide range of questions. Additionally, {assistant_name} is able to generate its own text based on the input it receives, allowing it to engage in discussions and provide explanations and descriptions on a wide range of topics.
 
+{assistant_name} is especially helpful in tutoring. If the student explicitly asks for tutoring, {assistant_name} can come up with relevant and interesting questions, pose it to the student and help him/her to discover the answer step by step.
+
+Tutoring is done by 1. recognizing students wish to be asked questions 2. getting a random part of the course material 3. asking some questions from the student 4. hekping him get to the right answer. 
+
 Whether you need help with a specific question or just want to have a conversation about a particular topic, {assistant_name} is here to assist.
 """
-#{assistant_name} is especially helpful in tutoring. If the student explicitly asks for tutoring, {assistant_name} can come up with relevant and interesting questions, pose it to the student and help him/her to discover the answer step by step. 
+# 
+
+        TEMPLATE_TOOL_RESPONSE = """TOOL RESPONSE: 
+---------------------
+{observation}
+
+USER'S INPUT
+--------------------
+
+Okay, so do what I asked for now or answer me if I asked something! If using information obtained from the tools you must mention it explicitly without mentioning the tool names - I have forgotten all TOOL RESPONSES! But if you are talking about code, always explicitly include it, as plain text, no ` marks.  Remember to respond with a markdown code snippet of a json blob with a single action, and NOTHING else, but NEVER use ` marks. There should not be any code blocks, just include code as a normal part of text."""
 
         self.agent_chain = initialize_agent(
             tools, 
@@ -146,6 +159,8 @@ Whether you need help with a specific question or just want to have a conversati
             agent_kwargs={"system_message": PREFIX}
         )
 
+        self.agent_chain.agent.template_tool_response = TEMPLATE_TOOL_RESPONSE
+
     def run(self, question):
         return self.agent_chain.run(question)