Congratulations!

Text-to-Query Agents with MongoDB and LangGraph

Apoorva Joshi

Senior AI/ML Developer Advocate, MongoDB

The finished product!

agentic10.jpg

Text-to-Query Agents with MongoDB and LangGraph

Your agent's toolbox

from langchain_mongodb.agent_toolkit.toolkit import MongoDBDatabaseToolkit

# Initialize the MongoDB database toolkit
toolkit = MongoDBDatabaseToolkit(db=db, llm=llm)

# Extract the tools from the toolkit
tools = toolkit.get_tools()
Text-to-Query Agents with MongoDB and LangGraph

Prompt engineering

from langchain_mongodb.agent_toolkit import MONGODB_AGENT_SYSTEM_PROMPT

# Create a templated prompt for the LLM
prompt = ChatPromptTemplate.from_messages(
    [
        ("system", MONGODB_AGENT_SYSTEM_PROMPT),
        ("system", "Do not re-run tools unless absolutely necessary. If you are not
                    able to get enough information using the tools, reply with
                    I DON'T KNOW. You have access to the following tools: {tool_names}."),
        MessagesPlaceholder(variable_name="messages"),
    ]
)
Text-to-Query Agents with MongoDB and LangGraph

Agent orchestration

langgraph.png

Text-to-Query Agents with MongoDB and LangGraph

Memory

mongosaver.png

thread_checkpoint.png

Text-to-Query Agents with MongoDB and LangGraph

What next?

 

More on MongoDB and Text-to-Query Agents

More on LangChain and LangGraph

Text-to-Query Agents with MongoDB and LangGraph

Congratulations!

Text-to-Query Agents with MongoDB and LangGraph

Preparing Video For Download...