State keys not updated in ChatPromptTemplate while using ChatPromptTemplate

[AkashBais]

I have constructed an agent as follows.

try:
        logger.info("Initializing agent creation")
        
        # Ensure the list_directory_contents tool is included in the tools list
        if list_directory_contents not in tools:
            tools.append(list_directory_contents)

        # Prepare tool names and team members as comma-separated strings
        tool_names = ', '.join([tool.name for tool in tools])
        team_members_str = ', '.join([str(member) for member in team_members])

        # Get the initial contents of the working directory
        # initial_directory_contents = list_directory_contents(path=working_directory)
        initial_directory_contents = list_directory_contents(working_directory)

        # Define the system prompt for the agent
        system_prompt = (
            "You are a specialized AI assistant in a data analysis team. "
            "Your role is to complete specific tasks in the research process. "
            "Use the provided tools to make progress on your task. "
            "If you can't fully complete a task, explain what you've done and what's needed next. "
            "Always aim for accurate and clear outputs. "
            f"You have access to the following tools: {tool_names}. "
            f"Your specific role: {system_message}\n"
            "Work autonomously according to your specialty, using the tools available to you. "
            "Do not ask for clarification. "
            "Your other team members (and other teams) will collaborate with you based on their specialties. "
            f"You are chosen for a reason! You are one of the following team members: {team_members_str}.\n"
            f"The initial contents of your working directory are:\n{initial_directory_contents}\n"
            "Use the ListDirectoryContents tool to check for updates in the directory contents when needed."
        )

        # Create the prompt template for the agent
        prompt = ChatPromptTemplate.from_messages(
            [
                ('system', system_prompt),
                MessagesPlaceholder(variable_name="messages"),
                ("ai", "hypothesis:{hypothesis}"),
                ("ai", "process:{process}"),
                ("ai", "process_decision:{process_decision}"),
                ("ai", "visualization_state:{visualization_state}"),
                ("ai", "search_state:{search_state}"),
                ("ai", "code_state:{code_state}"),
                ("ai", "report_section:{report_section}"),
                ("ai", "quality_review:{quality_review}"),
                ("ai", "needs_revision:{needs_revision}"),
                MessagesPlaceholder(variable_name="agent_scratchpad"),
            ]
        )

        # Create the agent with the specified model, tools, and prompt
        agent = create_react_agent(
            model=llm,
            tools=tools,
            state_modifier=prompt,
            debug=True,
        )

        logger.info("Agent creation completed successfully")
        return AgentExecutor.from_agent_and_tools(agent=agent,
                                                  tools=tools,
                                                  verbose=True).

This AgentExecutor is a node in a LangGraph graph. I am invoking the graph as

 graph.invoke( {"messages" : [message],
                      "visualization_state" : 'Initial visualization_state content',# AIMessage(content= 'Initial visualization_state content'),
                      "report_section" : 'Initial report_section content',#AIMessage(content= 'Initial report_section content'),
                      "needs_revision" : False,
                      "agent_scratchpad" : 'Initial agent_scratchpad content',#AIMessage(content= 'Initial agent_scratchpad content'),
                      "process" : 'Initial process content',#AIMessage(content= 'Initial process content'), 
                      "search_state" : 'Initial search_state content',#AIMessage(content= 'Initial search_state content'),
                      "code_state" : 'Initial code_state content',#AIMessage(content= 'Initial code_state content'),
                      "hypothesis" : 'Initial hypothesis content',#AIMessage(content= 'Initial hypothesis content'),
                      "quality_review" : 'Initial quality_review content',#AIMessage(content= 'Initial quality_review content'),
                      "process_decision" : 'Initial process_decision content',#AIMessage(content= 'Initial process_decision content'),
                       "agent_scratchpad": [],
              },
            {"configurable": {"thread_id": "42"}, "recursion_limit": 30} ,
            
            )

Every time I run the code it gives me an error as follows

2024-12-23 18:51:18,989 - core.Logger - INFO - Processing hypothesis_agent agent
2024-12-23 18:51:18,994 - core.Logger - INFO - State: {'messages': [HumanMessage(content='\nfile_path:hotels.csv\nUse machine learning to perform data analysis and write complete graphical reports\n', additional_kwargs={}, response_metadata={}, id='75d542e3-f9d4-49b3-8309-4595869a26b9')], 'process': 'Initial process content', 'hypothesis': 'Initial hypothesis content', 'process_decision': 'Initial process_decision content', 'visualization_state': 'Initial visualization_state content', 'search_state': 'Initial search_state content', 'code_state': 'Initial code_state content', 'report_section': 'Initial report_section content', 'quality_review': 'Initial quality_review content', 'needs_revision': False}
2024-12-23 18:51:19,007 - core.Logger - ERROR - Error while executing hypothesis_agent node: "Input to ChatPromptTemplate is missing variables {'needs_revision', 'hypothesis', 'visualization_state', 'process', 'report_section', 'search_state', 'code_state', 'quality_review', 'agent_scratchpad', 'process_decision'}.  Expected: ['agent_scratchpad', 'code_state', 'hypothesis', 'messages', 'needs_revision', 'process', 'process_decision', 'quality_review', 'report_section', 'search_state', 'visualization_state'] Received: ['messages', 'is_last_step', 'remaining_steps']\nNote: if you intended {needs_revision} to be part of the string and not a variable, please escape it with double curly braces like: '{{needs_revision}}'.\nFor troubleshooting, visit: https://python.langchain.com/docs/troubleshooting/errors/INVALID_PROMPT_INPUT "

The verbose logs from AgentExecutor.from_agent_and_tools are as follows.

> Entering new AgentExecutor chain...
[-1:checkpoint] State at the end of step -1:
{'messages': []}
[0:tasks] Starting 1 task for step 0:
- __start__ -> {'code_state': 'Initial code_state content',
 'hypothesis': 'Initial hypothesis content',
 'intermediate_steps': [],
 'messages': [HumanMessage(content='\nfile_path:hotels.csv\nUse machine learning to perform data analysis and write complete graphical reports\n', additional_kwargs={}, response_metadata={}, id='75d542e3-f9d4-49b3-8309-4595869a26b9')],
 'needs_revision': False,
 'process': 'Initial process content',
 'process_decision': 'Initial process_decision content',
 'quality_review': 'Initial quality_review content',
 'report_section': 'Initial report_section content',
 'search_state': 'Initial search_state content',
 'visualization_state': 'Initial visualization_state content'}
[0:writes] Finished step 0 with writes to 1 channel:
- messages -> [HumanMessage(content='\nfile_path:hotels.csv\nUse machine learning to perform data analysis and write complete graphical reports\n', additional_kwargs={}, response_metadata={}, id='75d542e3-f9d4-49b3-8309-4595869a26b9')]
[0:checkpoint] State at the end of step 0:
{'messages': [HumanMessage(content='\nfile_path:hotels.csv\nUse machine learning to perform data analysis and write complete graphical reports\n', additional_kwargs={}, response_metadata={}, id='75d542e3-f9d4-49b3-8309-4595869a26b9')]}
[1:tasks] Starting 1 task for step 1:
- agent -> {'is_last_step': False,
 'messages': [HumanMessage(content='\nfile_path:hotels.csv\nUse machine learning to perform data analysis and write complete graphical reports\n', additional_kwargs={}, response_metadata={}, id='75d542e3-f9d4-49b3-8309-4595869a26b9')],
 'remaining_steps': 29}

As seen while going from [0:tasks] Starting 1 task for step 0: to [0:writes] Finished step 0 with writes to 1 channel: all keys except messages are dropped. Can you halp me resolve the missing key error.

[dosubot[bot]]

I found a similar discussion that might help you resolve the issue with missing state keys in your ChatPromptTemplate. The error you're encountering is similar to a ValueError related to missing required variables in the prompt. In that discussion, the solution involved ensuring that the ChatPromptTemplate includes all necessary placeholders as MessagesPlaceholder entries ^[1].

Here's a potential solution based on that discussion:

1. Ensure all placeholders are correctly defined: Make sure that each variable you intend to use in the ChatPromptTemplate is defined as a placeholder. For example, if you have a variable like agent_scratchpad, it should be included as a MessagesPlaceholder.

2. Check the invocation inputs: Ensure that all required variables are being passed correctly when invoking the graph. The error message indicates that variables like needs_revision, hypothesis, visualization_state, etc., are missing. Double-check that these are included in the input dictionary when you call graph.invoke.

3. Review the state management: It seems that the state keys are not being maintained between steps. Ensure that your graph or agent logic is correctly managing and passing the state between nodes. This might involve checking how the state is updated and passed along in your graph's execution logic.

By ensuring that all placeholders are correctly defined and that the state is properly managed and passed, you should be able to resolve the missing key error.

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other}

[AkashBais]

@dosu The question you referred to doesn't answer the problem I have. As seen in the log trace below

2024-12-23 18:51:18,989 - core.Logger - INFO - Processing hypothesis_agent agent
2024-12-23 18:51:18,994 - core.Logger - INFO - State: {'messages': [HumanMessage(content='\nfile_path:hotels.csv\nUse machine learning to perform data analysis and write complete graphical reports\n', additional_kwargs={}, response_metadata={}, id='75d542e3-f9d4-49b3-8309-4595869a26b9')], 'process': 'Initial process content', 'hypothesis': 'Initial hypothesis content', 'process_decision': 'Initial process_decision content', 'visualization_state': 'Initial visualization_state content', 'search_state': 'Initial search_state content', 'code_state': 'Initial code_state content', 'report_section': 'Initial report_section content', 'quality_review': 'Initial quality_review content', 'needs_revision': False}
2024-12-23 18:51:19,007 - core.Logger - ERROR - Error while executing hypothesis_agent node: "Input to ChatPromptTemplate is missing variables {'needs_revision', 'hypothesis', 'visualization_state', 'process', 'report_section', 'search_state', 'code_state', 'quality_review', 'agent_scratchpad', 'process_decision'}.  Expected: ['agent_scratchpad', 'code_state', 'hypothesis', 'messages', 'needs_revision', 'process', 'process_decision', 'quality_review', 'report_section', 'search_state', 'visualization_state'] Received: ['messages', 'is_last_step', 'remaining_steps']\nNote: if you intended {needs_revision} to be part of the string and not a variable, please escape it with double curly braces like: '{{needs_revision}}'.\nFor troubleshooting, visit: https://python.langchain.com/docs/troubleshooting/errors/INVALID_PROMPT_INPUT "

2024-12-23 18:51:18,994 - core.Logger - INFO - State prints the content of the state just before invocation of the AgentExecutor.from_agent_and_tools abject and as seen it have all the required variables. Still why am I getting this missing keys error

[dosubot[bot]]

The error you're encountering suggests that the state keys are not being correctly passed to the ChatPromptTemplate during the execution of the AgentExecutor. Although the initial state seems to have all the required variables, they are not being maintained or passed correctly through the steps.

Here are a few things to check and try:

1. State Persistence: Ensure that the state is correctly persisted and passed between steps in your LangChain graph. The state should be updated and carried over from one node to the next. If the state is being reset or not properly updated, it could lead to missing keys.

2. State Management in Graph: Review how the state is managed within your graph. Make sure that each node in the graph correctly receives and updates the state. The state should be a shared object that is modified by each node as needed.

3. Input Dictionary: Double-check the input dictionary you are passing to graph.invoke. Ensure that all required variables are included and correctly formatted. The error message indicates that the expected variables are not being received, so there might be an issue with how the input is structured or passed.

4. Debugging Steps: Add additional logging to track the state as it moves through the graph. This can help identify where the state keys are being lost or not updated. Check the logs to see if the state is being modified unexpectedly between steps.

5. Agent Configuration: Verify that the AgentExecutor and ChatPromptTemplate are configured to handle the state correctly. Ensure that the placeholders in the ChatPromptTemplate match the keys in the state dictionary.

By ensuring that the state is correctly managed and passed through each step, you should be able to resolve the issue with missing keys.

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other}