> ## Documentation Index
> Fetch the complete documentation index at: https://docs.falkordb.com/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> FalkorDB is a graph database that speaks the Redis protocol. Queries are issued as OpenCypher through the GRAPH.QUERY and GRAPH.RO_QUERY commands, not over Bolt or a SQL connection.
> FalkorDB implements a subset of OpenCypher with proprietary extensions. Do not assume Neo4j-only syntax or procedures are available — check /cypher/cypher-support and /cypher/known-limitations before using a clause.
> FalkorDB is the successor to RedisGraph, but they are separate products. Do not present RedisGraph commands, versions, or limitations as current FalkorDB behavior.
> Use the official clients listed in /getting-started/clients rather than generic Redis or Neo4j drivers, and prefer the language the user is already working in.
> Configuration parameters are set with GRAPH.CONFIG SET or at startup; cite the exact parameter name from /getting-started/configuration rather than inventing one.
> This site covers four products: FalkorDB (core), FalkorDB Cloud, FalkorDB Enterprise, and the GraphRAG SDK. Name which one an answer applies to, since setup and operations differ.

# Agentic Memory

> Implement persistent, contextual memory for AI agents with FalkorDB using Graphiti, Cognee, and MCP servers for temporally-aware, multi-tenant knowledge graphs.

Agentic memory enables AI agents to maintain persistent, contextual memory across interactions. FalkorDB provides an ideal foundation for implementing agentic memory systems through its graph database capabilities, allowing agents to store, retrieve, and reason over complex relationships and temporal information.

## What is Agentic Memory?

Agentic memory refers to the ability of AI agents to:

* **Remember past interactions** and learn from them
* **Build contextual understanding** through connected knowledge
* **Reason over temporal information** to understand how relationships evolve
* **Share memory across agents** in multi-agent systems
* **Scale efficiently** as knowledge grows

## Why FalkorDB for Agentic Memory?

FalkorDB's graph database architecture makes it uniquely suited for agentic memory:

* **Graph-Native Storage**: Natural representation of entities, relationships, and contexts
* **Fast Traversals**: Quick retrieval of connected information for context-aware responses
* **Temporal Support**: Track how knowledge and relationships change over time
* **Multi-Tenant Architecture**: Isolated memory spaces for different agents or users
* **Hybrid Search**: Combine vector similarity with graph relationships for precise retrieval
* **High Performance**: Scale from prototype to production seamlessly

## Agentic Memory Frameworks

This section covers popular frameworks and tools that implement agentic memory with FalkorDB:

* [**Graphiti**](/agentic-memory/graphiti): A temporally-aware knowledge graph framework designed for multi-agent AI systems with persistent memory
* [**Graphiti MCP Server**](/agentic-memory/graphiti-mcp-server): Run Graphiti as an MCP server for Claude Desktop, Cursor IDE, and other AI clients *(Experimental)*
* [**Cognee**](/agentic-memory/cognee): A memory management framework for AI agents that combines graph and vector storage
* [**Mem0**](/agentic-memory/mem0): Add FalkorDB as a graph memory backend for Mem0 AI agents with per-user graph isolation

## Getting Started

Choose a framework based on your needs:

* If you need **temporal reasoning** and **multi-agent memory**, start with [Graphiti](/agentic-memory/graphiti)
* If you want to add **persistent memory to Claude Desktop or Cursor IDE**, try the [Graphiti MCP Server](/agentic-memory/graphiti-mcp-server)
* If you need **flexible memory structures** with **hybrid storage**, explore [Cognee](/agentic-memory/cognee)
* If you're using **Mem0 AI agents** and want **graph-structured memory**, integrate with [Mem0](/agentic-memory/mem0)

All frameworks integrate seamlessly with FalkorDB and can be used together in complex systems.

## Frequently Asked Questions

<AccordionGroup>
  <Accordion title="What is agentic memory and how does it differ from traditional database storage?">
    Agentic memory enables AI agents to maintain **persistent, contextual memory** across interactions. Unlike traditional storage, it uses graph structures to represent relationships between entities, supports temporal reasoning, and allows agents to build understanding over time rather than just storing flat records.
  </Accordion>

  <Accordion title="Why is a graph database like FalkorDB better suited for agentic memory than a relational database?">
    Graph databases naturally represent entities and their relationships, enabling fast traversals for context-aware retrieval. FalkorDB provides **native multi-graph support** for tenant isolation, hybrid search combining vectors with graph relationships, and temporal tracking - all essential for agentic memory systems.
  </Accordion>

  <Accordion title="Can I use multiple agentic memory frameworks together with FalkorDB?">
    Yes, all frameworks (Graphiti, Cognee, Mem0) integrate seamlessly with FalkorDB and can be used together in complex systems. Each framework can connect to the same FalkorDB instance using different graph names for isolation.
  </Accordion>

  <Accordion title="Which agentic memory framework should I choose?">
    Choose **Graphiti** for temporal reasoning and multi-agent memory, **Graphiti MCP Server** for adding memory to Claude Desktop or Cursor IDE, **Cognee** for flexible hybrid storage structures, or **Mem0** if you already use Mem0 agents and want graph-structured memory.
  </Accordion>

  <Accordion title="Do I need an LLM provider API key to use agentic memory with FalkorDB?">
    Yes, most agentic memory frameworks require an LLM provider (like OpenAI) for entity extraction and embedding generation. The LLM processes unstructured text into structured graph entities and relationships.
  </Accordion>
</AccordionGroup>