Claude Skill
MemTensor/MemOS
MemOS is a self-evolving memory OS for LLMs and AI agents, offering ultra-persistent memory, hybrid retrieval, and cross-task skill reuse with 35.24% token savings. Supports Claude Skill management...
Overview
Stars9,102
Forks815
LanguageTypeScript
Last pushed2026-05-15
Last synced2026-05-15
Repository
OwnerMemTensor
RepositoryMemOS
Full nameMemTensor/MemOS
Repo ID1,014,729,376
GitHub URLhttps://github.com/MemTensor/MemOS
🚀 Install this Skill
openclaw install MemTensor/MemOSSummary
MemOS is a self-evolving memory operating system designed for LLMs and AI agents, featuring ultra-persistent memory, hybrid retrieval, and cross-task skill reuse. It achieves 35.24% token savings and supports Claude Skill management for agentic AI workflows.
Chinese description
AI记忆操作系统，专为LLM与智能体系统设计，实现跨任务技能持久化存储、复用与进化能力的Claude Skill管理平台。
Key features
- Ultra-persistent memory for long-term agent context
- Hybrid retrieval combining semantic and structured search
- Cross-task skill reuse with self-evolving capabilities
- Claude Skill management platform for agentic AI
- 35.24% token savings through efficient memory optimization
- Self-hosted deployment with MCP support
Use cases
- Long-running AI agents requiring persistent context
- Multi-agent systems needing shared memory and skill reuse
- Token-efficient RAG pipelines for LLM applications
- Self-evolving AI workflows with adaptive memory management
- Claude Skill integration for agentic task automation