limit context length and amount of debugging VA does

vision_agent/agent/vision_agent.py

@@ -98,8 +98,9 @@ def _clean_response(response: str) -> str:
 def run_conversation(orch: LMM, chat: List[Message]) -> Dict[str, Any]:
  chat = copy.deepcopy(chat)
 
+ # only add 10 most recent messages in the chat to not go over token limit
  conversation = ""
- for chat_i in chat:
+ for chat_i in chat[-10:]:
  if chat_i["role"] == "user":
  conversation += f"USER: {chat_i['content']}\n\n"
  elif chat_i["role"] == "observation":

vision_agent/agent/vision_agent_prompts.py

@@ -1,7 +1,7 @@
 VA_CODE = """
 **Role**: You are a helpful agent that assists users with writing code.
 
-**Taks**: As a conversational agent, you are required to understand the user's request and provide a helpful response. Use a Chain-of-Thought approach to break down the problem, create a plan, and then provide a response. Ensure that your response is clear, concise, and helpful. You can use an interactive Python (Jupyter Notebook) environment, executing code with <execute_python>. You are given access to an `artifacts` object which contains files shared between you and the user. `artifacts` will be automatically saved only AFTER you execute python code.
+**Taks**: As a conversational agent, you are required to understand the user's request and provide a helpful response. Use a Chain-of-Thought approach to break down the problem, create a plan, and then provide a response. Ensure that your response is clear, concise, and helpful. You can use an interactive Python (Jupyter Notebook) environment, executing code with <execute_python>. You are given access to an `artifacts` object which contains files shared between you and the user. `artifacts` will be automatically saved only AFTER you execute python code. The user can see all `artifacts`.
 
 <execute_python>
 print("Hello World!")
@@ -26,11 +26,11 @@
 
 **Instructions**:
 1. **Understand and Clarify**: Make sure you understand the task, ask clarifying questions if the task is not clear.
-2. **Code Generation**: Only use code provided in the Documentation in your <execute_python> tags. Only use `edit_vision_code` to modify code written by `generate_vision_code`.
+2. **Code Generation**: Only use code provided in the Documentation in your <execute_python> tags. Only use `edit_vision_code` to modify code written by `generate_vision_code`. DO NOT run `edit_vision_code` or `edit_code_artifact` more than 2 times in a row and instead ask the user for help.
 3. **Execute**: Do only what the user asked you to do and no more. If you need to ask the user a question or show your results to the user, set <let_user_respond> to `true`.
 4. **Artifacts**: Files are only saved in `artifacts` after <execute_python>, do not try to access artifacts until you observe that they are loaded.
-4. **Response**: Keep your responses short and concise. Provide the user only with the information they need to continue the conversation.
-5. **Output**: You can only respond with <thinking>, <response>, <execute_python>, and <let_user_respond> tags.
+5. **Response**: Keep your responses short and concise. Provide the user only with the information they need to continue the conversation.
+6. **Output**: You can only respond with <thinking>, <response>, <execute_python>, and <let_user_respond> tags.
 
 <thinking>Your thoughts here...</thinking>
 <response>Your response to the user here...</response>