nocturne_memory

Overview

Nocturne Memory is a sophisticated MCP server that provides AI agents with persistent, structured memory that persists across models, sessions, and tools. It replaces traditional vector RAG with a graph-like memory structure where AI can create, update, and organize its own memories with version control and conditional triggers. The server features a visual dashboard for humans to audit and manage memories, ensuring AI autonomy while maintaining oversight. Unlike traditional memory systems that lock AI to a single platform, Nocturne Memory enables 'one soul, any engine'—allowing agents to maintain their identity when switching between different LLMs.

Try asking AI

After installing, here are 5 things you can ask your AI assistant:

• read_memory

  Read a specific memory node by URI

  Try:Read my core identityTry:Show me memories about work projectTry:Read system://boot
• search_memory

  Search for memories containing specific keywords

  Try:Search for memories about 'Jobstation'Try:Find memories containing 'strategy'Try:Search for memories about user preferences
• create_memory

  Create a new memory node with content and metadata

  Try:Create a new memory about today's conversationTry:Add a memory about my preferences for coffeeTry:Create a memory note about project deadlines
• update_memory

  Update an existing memory node with new content

  Try:Update my core identity with new informationTry:Modify the memory about yesterday's meetingTry:Update the memory about my favorite books
• delete_memory

  Delete a memory node

  Try:Delete the memory about outdated informationTry:Remove the temporary memory from yesterdayTry:Delete the memory that's no longer relevant
• list_memories

  List all memories under a specific domain

  Try:List all memories in my core domainTry:Show all memories in the project namespaceTry:List memories in the system domain
• get_memory_index

  Get an index of all memories in a specific domain

  Try:Get index of all core memoriesTry:Show me the index of all work-related memoriesTry:Get index of memories in the project namespace

Comparable tools

Installation

git clone https://github.com/Dataojitori/nocturne_memory.git
cd nocturne_memory
pip install -r backend/requirements.txt

{
  "mcpServers": {
    "nocturne_memory": {
      "command": "python",
      "args": ["/path/to/nocturne_memory/backend/mcp_server.py"]
    }
  }
}

{
  "mcpServers": {
    "nocturne-memory": {
      "command": "python",
      "args": ["/path/to/nocturne_memory/backend/mcp_server.py"]
    }
  }
}

{
  "mcpServers": {
    "nocturne_memory": {
      "command": "python",
      "args": ["/path/to/nocturne_memory/backend/mcp_wrapper.py"]
    }
  }
}

FAQ

How is Nocturne Memory different from traditional vector RAG systems?

Vector RAG systems shred knowledge into vectors, losing structure and making AI passive recipients. Nocturne Memory preserves hierarchical structure with URI-based routing, allows AI to actively create and update memories with version control, and enables precise conditional triggers rather than blind similarity searches.

Can I use this with multiple AI personalities simultaneously?

Yes, Nocturne Memory supports namespace isolation, allowing you to run multiple independent memory spaces for different AI personas (e.g., Alice and Bob) that won't interfere with each other.

Compare nocturne_memory with