deep-agents-memory
🎯Skillfrom langchain-ai/langchain-skills
A memory, persistence, and filesystem middleware skill for building agents with LangChain, LangGraph, and Deep Agents — installable via npx skills for any agent supporting the Agent Skills specification.
Overview
A memory, persistence, and filesystem middleware skill for building agents with LangChain, LangGraph, and Deep Agents. Part of the LangChain Skills collection, installable via npx for any agent supporting the Agent Skills specification including Claude Code, Cursor, Windsurf, and more.
Key Features
- Memory middleware — Persistent memory capabilities for long-running agent sessions
- Filesystem integration — File-based persistence for agent state and context
- LangGraph compatible — Works with LangGraph checkpointers and thread management
- Multi-agent install — Compatible with Claude Code, Cursor, Windsurf via
npx skills add - Deep Agents CLI support — Works with Deep Agents CLI including agent persona configuration
- Part of LangChain Skills — Alongside langgraph-persistence, langgraph-human-in-the-loop, and langchain-middleware skills
Who is this for?
Developers building LangChain or LangGraph agents who need persistent memory across sessions. Essential for teams creating long-running agents, multi-turn workflows, or applications where agent context must survive restarts and be shared across threads.
Same repository
langchain-ai/langchain-skills(25 items)
Installation
npx vibeindex add langchain-ai/langchain-skills --skill deep-agents-memorynpx skills add langchain-ai/langchain-skills --skill deep-agents-memory~/.claude/skills/deep-agents-memory/SKILL.mdSKILL.md
More from this repository10
LangGraph fundamentals skill from LangChain for building, observing, and evaluating agents with LangChain, LangGraph, and LangSmith.
A LangChain fundamentals skill that provides guidance for building, observing, and evaluating agents with LangChain, LangGraph, LangSmith, and Deep Agents.
Skill
A LangGraph persistence skill covering checkpointers, thread_id, and cross-thread memory — part of LangChain Skills for building agents with LangChain, LangGraph, and Deep Agents.
A LangGraph human-in-the-loop skill covering interrupts, human review, and approval workflows — part of LangChain Skills for building agents with LangChain, LangGraph, and Deep Agents.
Provides skills for orchestrating deep agents using LangChain, LangGraph, LangSmith, and Deep Agents, enabling building, observing, and evaluating multi-agent systems.
A Deep Agents core skill from LangChain Skills, providing patterns for building, observing, and evaluating agents with LangChain, LangGraph, LangSmith, and Deep Agents for AI coding workflows.
A LangChain middleware skill covering human-in-the-loop approval, custom middleware, and Command resume patterns — part of LangChain Skills for building agents with LangChain, LangGraph, and Deep Agents.
Agent skills for building, observing, and evaluating agents with LangChain, LangGraph, LangSmith, and Deep Agents, supporting both Claude Code and DeepAgents CLI installations.
A LangChain skill that helps select the right framework for building AI agents, covering LangChain, LangGraph, LangSmith, and Deep Agents.