KeyError: 'params'

[philiprhoades]

Describe the bug:
Problems with bot processing / retrieving info from the human text file.

Recreating VMs from scratch and / or:
Recreating humans and agents with:
SQL dump of DBs.
Clean of DBs.

Please describe your setup:
What is the output of memgpt version? (eg "0.2.4")
0.3.2 , 0.3.3 , 0.3.4, 0.3.7 for different VMs.

How did you install memgpt?

pip install pymemgpt
or:
pip install pymemgpt==0.3.2 etc

Describe your setup:
What's your OS:
Fedora Linux v38 (for Python3.11)

How are you running memgpt?
tmux Terminal

I am trying to use the changed CORE_MEMORY_PERSONA_CHAR_LIMIT and CORE_MEMORY_HUMAN_CHAR_LIMIT of 8000, but using all the defaults I still get the error below.

Thanks.

__ Logging login event.                                                                                                                            
__ Welcome back.                                                                                                                                   
> Enter your message: Can you tell me what you know about my siblings?                                                                             
An exception occurred when running agent.step():                                                                                                   
Traceback (most recent call last):                                                                                                                 
  File "/usr/local/lib/python3.11/site-packages/memgpt/local_llm/llm_chat_completion_wrappers/chatml.py", line 308, in output_to_chat_completion_re
sponse                                                                                                                                             
    function_parameters = function_json_output["params"]                                                                                           
                          ~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^                                                                                           
KeyError: 'params'                                                                                                                                 
                                                                                                                                                   
During handling of the above exception, another exception occurred:                                                                                
                                                                                                                                                   
Traceback (most recent call last):                                                                                                                 
  File "/usr/local/lib/python3.11/site-packages/memgpt/local_llm/chat_completion_proxy.py", line 171, in get_chat_completion                       
    chat_completion_result = llm_wrapper.output_to_chat_completion_response(result, first_message=first_message)                                   
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^                                   
  File "/usr/local/lib/python3.11/site-packages/memgpt/local_llm/llm_chat_completion_wrappers/chatml.py", line 310, in output_to_chat_completion_re
sponse                                                                                                                                             
    raise LLMJSONParsingError(                                                                                                                     
memgpt.errors.LLMJSONParsingError: Received valid JSON from LLM, but JSON was missing fields: 'params'. JSON result was:                           
{'function': 'camo_service', 'service_name': 'sibling_info', 'input_data': {'user_message': 'Can you tell me what you know about my siblings?'}, 'r
esponse_format': 'log_entry', 'time': '2024-03-26 03:54:52 AM UTC+0000'}                                                                           
                                                                                                                                                   
During handling of the above exception, another exception occurred:                                                                                
                                                                                                                                                   
Traceback (most recent call last):
  File "/usr/local/lib/python3.11/site-packages/memgpt/main.py", line 370, in run_agent_loop
    new_messages, user_message, skip_next_user_input = process_agent_step(user_message, no_verify)
                                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/site-packages/memgpt/main.py", line 346, in process_agent_step
    new_messages, heartbeat_request, function_failed, token_warning, tokens_accumulated = memgpt_agent.step(
                                                                                          ^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/site-packages/memgpt/agent.py", line 775, in step
    raise e
  File "/usr/local/lib/python3.11/site-packages/memgpt/agent.py", line 704, in step
    response = self._get_ai_reply(
               ^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/site-packages/memgpt/agent.py", line 413, in _get_ai_reply
    raise e
  File "/usr/local/lib/python3.11/site-packages/memgpt/agent.py", line 393, in _get_ai_reply
    response = create(
               ^^^^^^^
  File "/usr/local/lib/python3.11/site-packages/memgpt/llm_api_tools.py", line 379, in wrapper
    raise e
  File "/usr/local/lib/python3.11/site-packages/memgpt/llm_api_tools.py", line 352, in wrapper
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/site-packages/memgpt/llm_api_tools.py", line 471, in create
    return get_chat_completion(
           ^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/site-packages/memgpt/local_llm/chat_completion_proxy.py", line 176, in get_chat_completion
    raise LocalLLMError(f"Failed to parse JSON from local LLM response - error: {str(e)}")
memgpt.errors.LocalLLMError: Failed to parse JSON from local LLM response - error: Received valid JSON from LLM, but JSON was missing fields: 'params'. JSON result was:
{'function': 'camo_service', 'service_name': 'sibling_info', 'input_data': {'user_message': 'Can you tell me what you know about my siblings?'}, 'response_format': 'log_entry', 'time': '2024-03-26 03:54:52 AM UTC+0000'}
? Retry agent.step()? No
> Enter your message: /exit
Finished.

[philiprhoades]

This seems to be an "operator error" - I will close it.