skills.rest
victorlou
victorlouCOMMUNITY
Explore all skills by this author →
GitHub

agent-memory

Add long-term and short-term memory to LangGraph agents using Databricks Lakebase.

4|Updated May 9, 2026
Search Tags:#tooling#checkpointing#langgraph#agent-memory#state-persistence#lakebase#user-preferences
One-click install
npx skills add https://github.com/victorlou/housing-assistant --skill agent-memory-victorlou
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-memory
Source: https://github.com/victorlou/housing-assistant/tree/main/app/app-templates/.claude/skills/agent-langgraph-memory
Command: npx skills add https://github.com/victorlou/housing-assistant --skill agent-memory-victorlou

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires databricks-langchain[memory], databricks-sdk, langchain-core, langgraph-store, mlflow.

What problem does it solve?

This Skill helps you add persistent memory to an agent so user preferences and prior conversations can be reused instead of being forgotten between sessions.

Core Features & Use Cases

  • Long-term memory with Lakebase: Store and search user-specific facts (preferences, stable details) using a per-user namespace.
  • Conversation/session memory: Maintain short-term context within a thread via checkpointing to support multi-step exchanges.
  • Tooling for recall and updates: Provide agent tools to get, save, and delete memories with safe validation for JSON payloads.

Quick Start

Add the databricks-langchain[memory] dependency, configure Lakebase (including initializing tables with store.setup()), then integrate memory_tools() into your agent so it can recall and save user preferences using user_id.

Frequently Asked Questions about agent-memory

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I add long-term memory to a LangGraph agent for retaining user preferences?▼

To add long-term memory to a LangGraph agent, you integrate Databricks Lakebase (AsyncDatabricksStore) to store and search user-specific facts using a per-user namespace, preserving stable details across sessions.

What's the best way to maintain conversation context across sessions in a production agent?▼

Maintaining conversation context across sessions requires combining AsyncCheckpointSaver for short-term thread checkpointing with Lakebase for long-term storage, allowing the system to recall relevant history during natural-language interactions.

Does LangGraph support checkpointing with Databricks Lakebase for state persistence?▼

Yes, LangGraph supports state persistence by using AsyncCheckpointSaver for short-term checkpointing and Databricks Lakebase (AsyncDatabricksStore) for long-term memory, enabling multi-step exchanges within a thread.

How do I expose memory tools to an agent for getting and saving user-specific memories?▼

You expose memory tools by integrating the memory_tools() function into your agent, which provides tool-based functions for get, save, and delete operations with safe JSON validation and user_id extraction.

Do I need to initialize tables before using Databricks Lakebase for agent memory?▼

Yes, you need to initialize tables before using Databricks Lakebase by configuring the store and calling store.setup(), which prepares the database schema for retaining user preferences and conversation context.

Why does my agent forget user preferences between different requests?▼

Your agent forgets user preferences because it lacks persistent memory storage; adding Databricks Lakebase (AsyncDatabricksStore) allows the system to retain and recall stable user facts across requests and sessions.

Related Skills

Explore More Software Engineering Skills→
bcheng004bcheng004

agent-memory

Persist agent conversation checkpoints and user preferences via Databricks Lakebase.

Community
Intermediate
AshDaxAshDax

agent-memory

Persist user data across LangGraph agent sessions with Lakebase-backed memory tools.

Community
Advanced
databricksdatabricks
183

agent-memory

Integrate long-term and short-term memory into LangGraph agents with Databricks Lakebase.

Official
Advanced
thbehthbeh

databricks-model-serving

Deploy MLflow models and LangGraph AI agents to Databricks Model Serving endpoints.

Community
Advanced