OpenMemory and cortexgraph
These two tools are complements, as Cortexgraph could leverage OpenMemory as its local persistent memory store, with OpenMemory handling the lower-level key-value storage and persistence, while Cortexgraph provides the higher-level temporal memory logic and human-readable formatting.
About OpenMemory
CaviraOSS/OpenMemory
Local persistent memory store for LLM applications including claude desktop, github copilot, codex, antigravity, etc.
This project gives AI agents and large language models (LLMs) a persistent, long-term memory. It allows you to feed in information from various sources like GitHub, Notion, or web pages, and the AI can then recall and use these memories contextually over time. It's for developers building AI applications (e.g., chatbots, automated assistants, or intelligent UIs) who want their creations to remember past interactions and information without starting fresh every time.
About cortexgraph
prefrontal-systems/cortexgraph
Temporal memory system for AI assistants with human-like forgetting curves. All data stored locally in human-readable formats: JSONL for short-term memory, Markdown (Obsidian-compatible) for long-term. Memories naturally decay unless reinforced. Features knowledge graphs, smart prompting, and MCP server integration for Claude.
AI assistants often forget previous conversations, forcing you to repeat important information. CortexGraph helps your AI remember what matters, like your preferences or crucial facts, and naturally forget old, unused details over time. It takes your everyday conversations with an AI assistant and ensures key memories are reinforced and saved, much like human memory. This is for anyone who frequently interacts with AI assistants and wants them to retain context and preferences across sessions without constant retraining.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work