how to include history in few shot prompt with agent?

[octadion]

Checked other resources

• I added a very descriptive title to this question.
• I searched the LangChain documentation with the integrated search.
• I used the GitHub search to find a similar question and didn't find it.

Commit to Help

• I commit to help with one of those options 👆

Example Code

# prompt
def system_prompt():

    prompt_selector = SemanticSimilarityExampleSelector.from_examples(
        prompt,
        OpenAIEmbeddings(),
        FAISS,
        k=5,
        input_keys=["input"],
    )
    system_prefix = """You are an agent designed to interact with a SQL database.
    Given an input question in Indonesian language, create a syntactically correct {dialect} query to run, then look at the results of the query and return the answer.
    Unless the user specifies a specific number of examples they wish to obtain, always limit your query to at most {top_k} results.
    You can order the results by a relevant column to return the most interesting examples in the database.
    Never query for all the columns from a specific table, only ask for the relevant columns given the question.
    You have access to tools for interacting with the database.
    Only use the given tools. Only use the information returned by the tools to construct your final answer.
    You MUST double check your query before executing it. If you get an error while executing a query, rewrite the query and try again.
    
    DO NOT make any DML statements (INSERT, UPDATE, DELETE, DROP etc.) to the database.
    If the question does not seem related to the database, just return "I don't know" as the answer.
    Here are some examples of user inputs and their corresponding SQL queries:"""

    system_suffix = """Begin!
    Relevant pieces of previous conversation:
    {history}
    (Note: Only reference this information if it is relevant to the current query.)
    Question: {input}
    Thought: I should look at the tables in the database to see what I can query.  Then I should query the schema of the most relevant tables.
    {agent_scratchpad}
    """


    few_shot_prompt = FewShotPromptTemplate(
        example_selector=prompt_selector,
        example_prompt=PromptTemplate.from_template(
            "User input: {input}\nSQL query: {query}"
        ),
        input_variables=["input", "dialect", "top_k", "agent_scratchpad", "history"],
        prefix=system_prefix,
        suffix=system_suffix,
    )
    
    full_prompt = ChatPromptTemplate.from_messages(
        [
            MessagesPlaceholder(variable_name="history"),
            SystemMessagePromptTemplate(prompt=few_shot_prompt),
            ("human", "{input}"),
            MessagesPlaceholder("agent_scratchpad"),
        ]
    )

    return full_prompt

# agent
def create_agent(
        db,
        retriever_tool=None, 
        function_tool=None,
        llm=str
    ):

    llm_agent = get_agent_llm(llm)
    memory = ConversationBufferMemory(memory_key="history", input_key="input")
    
    extra_tools = []
    if retriever_tool is not None:
        extra_tools.append(retriever_tool)
    if function_tool is not None:
        extra_tools += function_tool

    agent = create_sql_agent(
        llm=llm_agent,
        db=db,
        agent_type="openai-tools",
        prompt=full_prompt,
        extra_tools=extra_tools,
        input_variables=["input", "agent_scratchpad", "history"],
        verbose=True,
        agent_executor_kwargs={"memory": memory, "handle_parsing_errors":True},
    )
    return agent

Description

i want to include history on my agent so it can answer based on previous conversations, but based on the code I created, I get an error:
ValueError: variable history should be a list of base messages, got
Any idea? thank you!

System Info

"pip freeze"
platform (windows)
python version (3.0)
langchain:
langchain==0.1.13
langchain-community==0.0.29
langchain-core==0.1.34
langchain-openai==0.1.1
langchain-text-splitters==0.0.1
langsmith==0.1.33

[dosubot[bot]]

The error you're encountering suggests there's a mismatch in the expected format for the history variable. It should be a list of BaseMessage instances, but it seems the format you're passing is not recognized as such. This issue often arises from how the conversation history is managed, particularly in how messages are stored, retrieved, and formatted before being passed to the agent.

To address this, ensure that:

• Your conversation history retrieval process correctly returns a list of BaseMessage instances.
• Messages are properly serialized and deserialized when storing and retrieving from your storage solution, maintaining the correct format.
• Your application correctly handles session management, ensuring that the correct conversation history is retrieved for each session.

Additionally, you mentioned using Python 3.0, but LangChain requires Python version ">=3.8.1,<4.0". Upgrading your Python version to at least 3.8.1 might resolve some compatibility issues.

If these steps don't resolve the issue, please check how you're storing and retrieving the conversation history, including the serialization and deserialization process, and ensure it aligns with the expected BaseMessage format.

Details

Sources

• multi chat.invoke and RunnableWithMessageHistory pass parameter fail
• Adding examples in the Conversation Chain along with maintaining history
• libs/core/langchain_core/runnables/history.py
• libs/partners/mongodb/langchain_mongodb/chat_message_histories.py
• libs/langchain/langchain/memory/buffer.py
• libs/langchain/langchain/chains/history_aware_retriever.py
• libs/langchain/pyproject.toml
• libs/community/langchain_community/utilities/init.py
• docs/docs/contributing/integrations.mdx
• libs/langchain/langchain/agents/load_tools.py

About Dosu
This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

To continue the conversation, mention @dosu.

[jakelcoop]

@octadion Did you ever figure this out?

[dotmatss]

@octadion did you manage to add?