LangChain Tools + MemorySync
Structured memory tools that let LangChain agents store, search, list, update, and delete memories safely.
Overview
One factory call returns five structured tools — add_memory, search_memory, list_memories, update_memory, delete_memory — with identical names and behaviour in Python and Node.js. Tools return readable strings and never raise, so a memory failure cannot abort an agent run, and writes are idempotent so a repeating agent cannot create duplicates.
Setup status and requirements
- Supported version
- langchain-memorysync 1.1.0 (PyPI) / memorysync-langchain 1.1.0 (npm)
- Last setup review
- 2026-08-22
- Permissions
- A MemorySync API key with read and write scopes; delete_memory needs delete permission.
- Limits
- Requires langchain-core 0.3 or newer (Python 3.10+, Node 18+). Memory text is immutable — update changes tags and importance, and replacing a fact is delete plus add.
Capabilities
- Five tools from one factory call: add, search, list, update, delete
- Tools never raise — failures come back as readable strings
- Content-derived idempotency: repeated saves are recognised, not duplicated
- Per-end-user scoping — a delete cannot reach another user’s memory
- read_only mode returning only search and list
- Works with create_agent, LangGraph agents, and the legacy AgentExecutor
Quick Start
from langchain.agents import create_agentfrom langchain_memorysync import create_memorysync_toolstools = create_memorysync_tools(end_user_id="customer-7")agent = create_agent(model, tools=tools)
Use Cases
Agents that decide what to remember
Explicit store-and-recall workflows
Read-only memory access for untrusted agents
Explore More
Built for production AI systems
Build AI systems that remember
MemorySync provides the infrastructure layer for persistent memory, adaptive retrieval, and enterprise AI intelligence.
Start free. No credit card required.