Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.maximem.ai/llms.txt

Use this file to discover all available pages before exploring further.

​
Install

pip install maximem-synap-langchain

​
What’s included

synap-langchain provides four drop-in components:
ClassPurpose
SynapChatMessageHistoryPersistent chat history via BaseChatMessageHistory
SynapCallbackHandlerAuto-records every LLM turn via LangChain callbacks
SynapRetrieverSemantic retriever that returns Synap memories as Document objects
SynapSearchTool, SynapStoreToolAgent-callable tools for explicit memory read/write

​
SynapChatMessageHistory

Plug into RunnableWithMessageHistory to give any chain persistent memory across sessions. 
from langchain_core.runnables.history import RunnableWithMessageHistory
from synap_langchain import SynapChatMessageHistory

def get_history(session_id: str):
    return SynapChatMessageHistory(
        sdk=sdk,
        conversation_id=session_id,
        user_id="alice",
        customer_id="acme",   # optional
    )

chain_with_history = RunnableWithMessageHistory(
    base_chain,
    get_session_history=get_history,
)

response = await chain_with_history.ainvoke(
    {"question": "What did we discuss last time?"},
    config={"configurable": {"session_id": "conv-123"}},
)

​
SynapCallbackHandler

Attach to any chain or agent to automatically ingest every conversation turn without changing your application logic. 
from synap_langchain import SynapCallbackHandler

handler = SynapCallbackHandler(
    sdk=sdk,
    conversation_id="conv-123",
    user_id="alice",
)

response = await chain.ainvoke(
    {"question": "Remind me of my project deadlines."},
    config={"callbacks": [handler]},
)
Failures during ingestion are logged at ERROR level and do not raise — your chain always completes.

​
SynapRetriever

Use as a standard LangChain retriever in RAG pipelines or ConversationalRetrievalChain. 
from synap_langchain import SynapRetriever

retriever = SynapRetriever(
    sdk=sdk,
    user_id="alice",
    customer_id="acme",
    max_results=8,
    mode="fast",    # or "accurate"
)

docs = await retriever.aget_relevant_documents("project deadlines")
# Each doc: page_content = memory text, metadata = {"confidence": 0.92, "type": "fact", ...}

​
SynapSearchTool and SynapStoreTool

Give agents explicit control over memory read and write. 
from langchain.agents import AgentExecutor, create_tool_calling_agent
from synap_langchain import SynapSearchTool, SynapStoreTool

tools = [
    SynapSearchTool(sdk=sdk, user_id="alice", customer_id="acme"),
    SynapStoreTool(sdk=sdk, user_id="alice", customer_id="acme"),
]

agent = create_tool_calling_agent(llm, tools, prompt)
executor = AgentExecutor(agent=agent, tools=tools)

​
Next steps

LangGraph

Checkpointer and store for LangGraph state graphs.

SDK Ingestion

Direct ingestion API for custom pipelines.