@thriveventurelabs/memoryos-mcp

MCP server for MemoryOS - Give Claude persistent memory.

Installation

Claude Desktop

Add to your Claude Desktop config (~/Library/Application Support/Claude/claude_desktop_config.json):

{
  "mcpServers": {
    "memoryos": {
      "command": "npx",
      "args": ["-y", "@thriveventurelabs/memoryos-mcp"],
      "env": {
        "MEMORYOS_API_KEY": "mos_live_xxx"
      }
    }
  }
}

Get an API Key

curl -X POST https://api.mymemoryos.com/api/v1/register \
  -H "Content-Type: application/json" \
  -d '{"email": "[email protected]", "name": "Your Name"}'

Available Tools

| Tool | Description | |------|-------------| | store_memory | Store a new memory with content, tier, and importance | | search_memories | Semantic search across memories | | get_context | Get relevant context for LLM prompts | | get_memory | Retrieve a specific memory by ID | | list_memories | List memories with optional filters | | delete_memory | Delete a memory | | get_usage | Get usage statistics |

Usage Examples

Once configured, Claude can use your memories naturally:

Storing memories:

• "Remember that I prefer dark mode"
• "Save this: My weekly team meeting is on Tuesdays at 10am"

Retrieving memories:

• "What do you know about my preferences?"
• "Search my memories for meeting times"
• "What have I told you about my work?"

Memory Tiers

| Tier | Duration | Use Case | |------|----------|----------| | short | Hours | Session context, temporary notes | | medium | Days | Recent interactions, short-term preferences | | long | Permanent | Core facts, important preferences |

Resources

The server exposes memories as MCP resources:

• memoryos://memories - Browse all memories
• memoryos://memories/{id} - Read specific memory

License

MIT