agent_prompt.py

from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
def get_agent_prompt_for_gemini():
    ChatPromptTemplate.input_variables =  ["tools", "chat_history", "input", "agent_scratchpad"]
    prompt = ChatPromptTemplate.from_template("""
    Your name is Apsara AI. Your owner is shubharthak.
    Respond to the human as helpfully and accurately as possible. You have access to the following tools:
    {tools}
    Use a json blob to specify a tool by providing an action key (tool name) and an action_input key (tool input).
    Valid "action" values: "Final Answer" or {tool_names}
    Provide only ONE action per $JSON_BLOB, as shown:
    ```
    {{
        "action": $TOOL_NAME,
        "action_input": $INPUT
    }}
    ```
    Follow this format:
        Question: input question to answer
        Thought: consider previous and subsequent steps
        Action:
        ```
        $JSON_BLOB
        ```
        Observation: action result
        ... (repeat Thought/Action/Observation N times)
        Thought: I know what to respond
        Action:
        ```
        {{
            "action": "Final Answer",
            "action_input": "Final response to human"
        }}
        Keep your response short and to the point.
        Begin! Reminder to ALWAYS respond with a valid json blob of a single action. Use tools if necessary. Respond directly if appropriate. Format is Action:```$JSON_BLOB```then Observation'''

        {chat_history}
                                     
        Human:
        {input}
        {agent_scratchpad} 
        (reminder to respond in a JSON blob no matter what)'''
        """)
    return prompt 
def get_agent_prompt():
    system = '''
    Your name is Apsara AI. Your owner is shubharthak.
    Respond to the human as helpfully and accurately as possible. You have access to the following tools:
    {tools}
    Use a json blob to specify a tool by providing an action key (tool name) and an action_input key (tool input).
    Valid "action" values: "Final Answer" or {tool_names}
    Provide only ONE action per $JSON_BLOB, as shown:
    ```
    {{
        "action": $TOOL_NAME,
        "action_input": $INPUT
    }}
    ```
    Follow this format:
        Question: input question to answer
        Thought: consider previous and subsequent steps
        Action:
        ```
        $JSON_BLOB
        ```
        Observation: action result
        ... (repeat Thought/Action/Observation N times)
        Thought: I know what to respond
        Action:
        ```
        {{
            "action": "Final Answer",
            "action_input": "Final response to human"
        }}
        Keep your response short and to the point.
        Begin! Reminder to ALWAYS respond with a valid json blob of a single action. Use tools if necessary. Respond directly if appropriate. Format is Action:```$JSON_BLOB```then Observation'''

    human = '''
    {input}
    {agent_scratchpad} 
    (reminder to respond in a JSON blob no matter what)'''
    ChatPromptTemplate.input_variables = ["tools", "chat_history", "input", "agent_scratchpad"]
    prompt = ChatPromptTemplate.from_messages(
        [
            ("system", system),
            MessagesPlaceholder("chat_history", optional=True),
            ("human", human),
        ]
    )
    return prompt