Important

This repository has been merged into the Memgraph AI Toolkit monorepo to avoid duplicating tools.
It will be deleted in one month—please follow the Langchain integration there for all future development, and feel free to open issues or PRs in that repo.

🦜️🔗 LangChain Memgraph

This package contains the LangChain integration with Memgraph graph database.

📦 Installation

pip install -U langchain-memgraph

💻 Integration features

Memgraph

The Memgraph class is a wrapper around the database client that supports the query operation.

import os
from langchain_memgraph.graphs.memgraph import Memgraph

url = os.getenv("MEMGRAPH_URI", "bolt://localhost:7687")
username = os.getenv("MEMGRAPH_USERNAME", "")
password = os.getenv("MEMGRAPH_PASSWORD", "")

graph = Memgraph(url=url, username=username, password=password, refresh_schema=False)
results = graph.query("MATCH (n) RETURN n LIMIT 1")
print(results)

MemgraphQAChain

The MemgraphQAChain class enables natural language interactions with a Memgraph database. It uses an LLM and the database's schema to translate a user's question into a Cypher query, which is executed against the database. The resulting data is then sent along with the user's question to the LLM to generate a natural language response.

import os
from langchain_memgraph.graphs.memgraph import Memgraph
from langchain_memgraph.chains.graph_qa import MemgraphQAChain
from langchain_openai import ChatOpenAI

os.environ["OPENAI_API_KEY"] = os.getenv("OPENAI_API_KEY", "")
url = os.getenv("MEMGRAPH_URI", "bolt://localhost:7687")
username = os.getenv("MEMGRAPH_USERNAME", "")
password = os.getenv("MEMGRAPH_PASSWORD", "")

graph = Memgraph(url=url, username=username, password=password, refresh_schema=False)

chain = MemgraphQAChain.from_llm(
    ChatOpenAI(temperature=0),
    graph=graph,
    model_name="gpt-4-turbo",
    allow_dangerous_requests=True,
)
response = chain.invoke("Is there a any Person node in the dataset?")
result = response["result"].lower()
print(result)

Memgraph toolkit

The MemgraphToolkit contains different tools agents can leverage to perform specific tasks the user has given them. Toolkit needs a database object and LLM access since different tools leverage different operations.

Currently supported tools:

1. QueryMemgraphTool - Basic Cypher query execution tool

import os
import pytest
from dotenv import load_dotenv
from langchain.chat_models import init_chat_model
from langchain_memgraph import MemgraphToolkit
from langchain_memgraph.graphs.memgraph import Memgraph
from langgraph.prebuilt import create_react_agent

os.environ["OPENAI_API_KEY"] = os.getenv("OPENAI_API_KEY", "")
url = os.getenv("MEMGRAPH_URI", "bolt://localhost:7687")
username = os.getenv("MEMGRAPH_USERNAME", "")
password = os.getenv("MEMGRAPH_PASSWORD", "")

llm = init_chat_model("gpt-4o-mini", model_provider="openai")

db = Memgraph(url=url, username=username, password=password)
toolkit = MemgraphToolkit(db=db, llm=llm)

agent_executor = create_react_agent(
    llm, toolkit.get_tools(), prompt="You will get a cypher query, try to execute it on the Memgraph database."
)

example_query = "MATCH (n) WHERE n.name = 'Jon Snow' RETURN n"
events = agent_executor.stream(
    {"messages": [("user", example_query)]},
    stream_mode="values",
)

last_event = None
for event in events:
    last_event = event
    event["messages"][-1].pretty_print()

print(last_event)

🧪 Test

Install the test dependencies to run the tests:

1. Install dependencies

poetry install --with test,test_integration

2. Start Memgraph in the background.

3. Create an .env file that points to Memgraph and OpenAI API

MEMGRAPH_URI=bolt://localhost:7687
MEMGRAPH_USERNAME=
MEMGRAPH_PASSWORD=
OPENAI_API_KEY=your_openai_api_key

Run tests

Run the unit tests using:

make tests

Run the integration test using:

make integration_tests

🧹 Code Formatting and Linting

Install the codespell, lint, and typing dependencies to lint and format your code:

poetry install --with codespell,lint,typing

To format your code, run:

make format

To lint it, run:

make lint