Agent♥︎Age
Catalog

Montycat MCP - Shared Memory for AI Agents

Official

by MontyGovernance · Python

Shared, persistent memory for AI agents, with semantic recall and live updates.

io.github.MontyGovernance/montycat-mcp (MCP Server)

Montycat MCP is a local-first Model Context Protocol (MCP) memory server that provides shared, persistent memory for AI agents. It supports semantic recall and live updates. The server targets tools including Claude, OpenAI Codex, Cursor, and “”.

🛠️ Key Features

  • Shared memory for AI agents
  • Persistent memory
  • Semantic recall
  • Live updates
  • Implements MCP (“Model Context Protocol”)

🚀 Use Cases

  • Remembering information across agent interactions
  • Sharing state between multiple agents
  • Semantic retrieval for agent knowledge (via recall)

⚡ Developer Benefits

  • Fits into MCP-based setups (model-context-protocol)
  • Supports multi-agent workflows with shared-memory
  • Uses vector/embedding-related capabilities (embeddings, vector-database, vector-search)

⚠️ Limitations

  • Available description excerpt is incomplete and does not fully specify supported clients beyond “Claude, OpenAI Codex, Cursor, and …”.

Topics

agent-memoryai-memoryclaudeclaude-codecodexcursorembeddingsllmlong-term-memorymcpmcp-servermodel-context-protocolmulti-agentpersistent-memoryragself-hostedsemantic-searchshared-memoryvector-databasevector-search