Langgraph Persistence

Memory that actually works in LangGraph applications. This covers the checkpointer system that saves graph state at every step, thread management for isolated conversations, and the Store API for cross-thread data like user preferences. Shows you how to set up PostgreSQL persistence for production, navigate checkpoint history for debugging, and handle the tricky subgraph scoping decisions that determine whether child graphs inherit persistence or run isolated. The subgraph checkpointer table alone saves hours of trial and error when building complex multi-agent workflows that need human-in-the-loop interrupts or stateful conversations.

skills.sh installs: 18.8K
GitHub stars: 1.3K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add langchain-ai/langchain-skills --skill langgraph-persistence --agent claude-code

Installs into .claude/skills of the current project.

Files

LangGraph's persistence layer enables durable execution by checkpointing graph state:

Checkpointer: Saves/loads graph state at every super-step
Thread ID: Identifies separate checkpoint sequences (conversations)
Store: Cross-thread memory for user preferences, facts

Two memory types:

Short-term (checkpointer): Thread-scoped conversation history
Long-term (store): Cross-thread user preferences, facts

Checkpointer	Use Case	Production Ready
`InMemorySaver`	Testing, development	No
`SqliteSaver`	Local development	Partial
`PostgresSaver`	Production	Yes

Checkpointer Setup

Set up a basic graph with in-memory checkpointing and thread-based state persistence.

from langgraph.checkpoint.memory import InMemorySaver
from langgraph.graph import StateGraph, START, END
from typing_extensions import TypedDict, Annotated
import operator

class State(TypedDict):
    messages: Annotated[list, operator.add]

SKILL.md on GitHub