MemTensor — 5 Open-Source AI Agent Skills

MemTensor is the author of 5 open-source AI agent skills and MCP servers spanning mcp-server, codex-skill, agent-tool, with a combined 14,110+ GitHub stars. On AgentSkillsHub each is quality-scored (avg 68/100) and security-graded — 5 verified safe.

Top skills by MemTensor

• MemOS — Self-evolving memory OS for LLM & AI Agents: ultra-persistent memory, hybrid-retrieval, and cross-task skill reuse, with 35.24% token saving (11.5k★ · quality 62/100 · security: safe)
• memmy-agent — 🍙 A personal AI agent & local memory hub for all AI agents, gives every AI one shared, fully controlled memory and persistent context — al (2.0k★ · quality 65/100 · security: safe)
• MemOS-Cloud-OpenClaw-Plugin — Official MemOS Cloud plugin for OpenClaw. Enables long-term memory for agents by recalling context before execution and saving conversations (370★ · quality 74/100 · security: safe)
• skills-vote — SkillsVote: Lifecycle Governance of Agent Skills from Collection, Recommendation to Evolution (302★ · quality 67/100 · security: safe)
• memos-api-mcp — A Model Context Protocol (MCP) implementation for the Memos API service. This package provides a command-line interface to interact with Mem (7★ · quality 74/100 · security: safe)

View MemTensor on GitHub · Explore audited agent skills