DesktopCommanderMCP plus OpenMemory MCP?

[dts350z]

Seems like it would be a powerful combo.

https://levelup.gitconnected.com/how-to-make-your-clients-more-context-aware-with-openmemory-mcp-60057bcc24a3

but one question would be would it eat up tokens even faster? That seems (running our of tokens in a conversation) to be the bigfeet limitation that keeps sending me back to Windsurf, but on the other hand, maybe this enables you to keep context across conversations so actually enable more Claude + DesktopCommanderMCP usage?

[dts350z]

That particular implementation seems overly complicated:

Why do you need a "frontend" when the goal is an MCP server?
Why do you need an LLM api key, when the goal is an MCP server?

but the concept off adding memory to DesktopCommanderMCP sounds (really) good (hint hint).

[serg33v]

Memory as a concept is really interesting, but making it work as expected is difficult because there’s no clear definition of how it should function. Feel free to share your opinion — we’re collecting feedback from our users and will implement something that’s genuinely helpful.
Personally, I think memory is a consequence of the small context window. If the context size were 10 million tokens, memory wouldn’t be necessary.

[dts350z]

I said:
I'm starting to see MCP server projects to add memories to LLMs. The specific one mentioned in an article I read seems overly complicated: https://github.com/mem0ai/mem0/tree/main/openmemory?ref=mem0.ai

I'm wondering if there is a less complex solution (no frontend needed, no LLM Api key needed, etc.).?

The ultimate goal is code in Claude + DeskTopCommanderMCP and keep context and memories across Claude conversations.

Claude 4 Sonnet said:

You're absolutely right - the mem0ai OpenMemory setup is overly complex for your needs. It requires Docker, PostgreSQL, Qdrant vector database, a web UI, and OpenAI API keys. That's a lot of infrastructure for what should be simple persistent memory.

Here's a much simpler solution: Basic Memory by Basic Machines. https://github.com/basicmachines-co/basic-memory

Why Basic Memory is Perfect for Your Use Case
What makes it simple:

No Docker, databases, or complex setup - just local Markdown files GitHubPyPI
No LLM API keys required PlaybooksGitHub
No frontend UI needed Basic Memory MCP server for AI agents
Lightweight infrastructure: Just local files indexed in a local SQLite database GitHub - basicmachines-co/basic-memory: Basic Memory is a knowledge management system that allows you to build a persistent semantic graph from conversations with AI assistants, stored in standard Markdown files on your computer. Integrates directly with Obsidan.md

How it works:

Stores knowledge in standard Markdown files on your computer GitHubPyPI
Uses the Model Context Protocol (MCP) to enable Claude to read and write to your local knowledge base PlaybooksGitHub
Creates persistent semantic graphs from conversations GitHub - basicmachines-co/basic-memory: Basic Memory is a knowledge management system that allows you to build a persistent semantic graph from conversations with AI assistants, stored in standard Markdown files on your computer. Integrates directly with Obsidan.md
Everything stays local - you have full control and ownership GitHub - basicmachines-co/basic-memory: Basic Memory is a knowledge management system that allows you to build a persistent semantic graph from conversations with AI assistants, stored in standard Markdown files on your computer. Integrates directly with Obsidan.md

Installation
For Claude Desktop:
json{
"mcpServers": {
"basic-memory": {
"command": "uvx",
"args": ["basic-memory", "mcp"]
}
}
}
Or use the automated installer:

npx -y @smithery/cli install @basicmachines-co/basic-memory --client claude

How It Works in Practice
When you chat with Claude, it can:

Save information: Create new Markdown files for topics you discuss GitHub - basicmachines-co/basic-memory: Basic Memory is a knowledge management system that allows you to build a persistent semantic graph from conversations with AI assistants, stored in standard Markdown files on your computer. Integrates directly with Obsidan.md
Retrieve context: Read from your knowledge base to maintain context between conversations Basic Memory MCP server for AI agents
Build relationships: Links related concepts together GitHub - basicmachines-co/basic-memory: Basic Memory is a knowledge management system that allows you to build a persistent semantic graph from conversations with AI assistants, stored in standard Markdown files on your computer. Integrates directly with Obsidan.md

The files are stored with YAML frontmatter and follow a specific format GitHub - basicmachines-co/basic-memory: Basic Memory is a knowledge management system that allows you to build a persistent semantic graph from conversations with AI assistants, stored in standard Markdown files on your computer. Integrates directly with Obsidan.md, but you can view and edit them directly in any text editor or Obsidian.

Comparison to mem0ai OpenMemory

Featuremem0ai OpenMemoryBasic MemorySetup complexityHigh (Docker, PostgreSQL, Qdrant)Low (single command)LLM API requiredYes (OpenAI)NoStorageVector databaseLocal Markdown filesUIYes (web interface)No (file-based)DependenciesManyMinimal

This should give you exactly what you want: persistent memory across Claude conversations using DesktopCommanderMCP, without the complexity of the mem0ai solution.