Agent♥︎Age
Catalog

Living Memory

OfficialLive

by v1b3x0r · TypeScript

An MCP memory server. One memory your agents share — across models, devices and apps.

to.viibe/living-memory MCP Memory Server

This MCP server is a “Living Memory” memory server that provides one shared memory for agents across models, devices, and apps. It is associated with a “Living Memory Engine” and is published under @nature-labs/lme-mcp (MCP server package) and @nature-labs/living-memory-engine (engine package).

🛠️ Key Features

  • MCP memory server
  • One shared memory for agents
  • Memory shared across models, devices, and apps
  • Topics include local-first, IndexedDB, vector-search-embeddings, and RAG

🚀 Use Cases

  • Agent memory that persists across environments (models/devices/apps)
  • Retrieval-oriented workflows using vector search and embeddings
  • Local-first storage via IndexedDB

⚡ Developer Benefits

  • Works within MCP (mcp, mcp-server)
  • Use cases align with RAG (rag) and vector search (vector-search-embeddings)
  • Implemented in TypeScript (typescript)

⚠️ Limitations

  • Source material only identifies the server as an MCP memory service; no specific tool interfaces, APIs, or constraints are provided here.

Topics

ai-agentsebbinghaus-decayindexeddblocal-firstmemory-engineollamaqwenragtypescriptvector-search-embeddingschatgptclaude-codecursorhermes-agentmcpmcp-server