memhq-langchain 0.1.0

MemHQ adapter for LangChain — chat message history and retriever backed by MemHQ.

memhq-langchain

MemHQ adapter for LangChain (Python). Drop-in chat message history and retriever backed by the MemHQ hosted API. Sync and async both supported via httpx.

Install

pip install memhq-langchain

Quickstart — Chat history

import os
from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
from langchain_core.runnables.history import RunnableWithMessageHistory

from memhq_langchain import MemHQChatMessageHistory

model = ChatOpenAI(model="gpt-4o-mini")
prompt = ChatPromptTemplate.from_messages([
    ("system", "You have access to the user's memory."),
    MessagesPlaceholder("history"),
    ("human", "{input}"),
])

chain = RunnableWithMessageHistory(
    prompt | model,
    lambda session_id: MemHQChatMessageHistory(
        api_key=os.environ["MEMHQ_API_KEY"],
        session_id=session_id,
        user_id="user_42",
    ),
    input_messages_key="input",
    history_messages_key="history",
)

reply = chain.invoke(
    {"input": "What's my favorite coffee order?"},
    config={"configurable": {"session_id": "conv_42"}},
)

Quickstart — Retriever

from memhq_langchain import MemHQRetriever

retriever = MemHQRetriever(
    api_key=os.environ["MEMHQ_API_KEY"],
    user_id="user_42",
    limit=8,
)

docs = retriever.invoke("pricing discussions")

Configuration

Argument	Description
api_key	MemHQ project API key. Falls back to MEMHQ_API_KEY.
user_id	External user id. Required.
session_id	Thread anchor for the history adapter.
recall_limit	Memories surfaced per history read. Default 10.
limit	Memories returned per retriever query. Default 10.
mode	"hybrid" (default), "vector", or "lexical".
base_url	Override the API base.

Reference

Full reference: https://docs.memhq.ai/sdks/langchain-py