GitHub Repository

Persistent local memory for Claude and ChatGPT. No cloud, no subscription. Self-hosted RAG over MCP.

4 starsPython

Local RAG memory system that AI can write directly to

Name: Local RAG memory system that AI can write directly to
Availability: InStock
Author: ptobey

by ptobey·Jun 13, 2026·3 points·1 comment

Visit Project View on HN

AI Analysis

●●SolidSolve My ProblemShip It

Reconciliation and versioning beat basic MCP memory servers that just store chunks.

Strengths

•Heuristic reconciliation detects conflicts and returns self-heal hints on risky writes.
•Versioned updates with supersedes chains keep memory evolution tracked properly.
•Stdio and SSE transports with optional bearer or OAuth auth for deployment flexibility.

Weaknesses

•Local AI memory MCP servers are becoming a crowded category with multiple options.
•Only 4 stars suggests limited real-world testing beyond the author's family use.

Post Description

For me and my family, it get really annoying having to reshare information each time you create a new LLM chat.

Therefore, I decided to create local-memory-mcp, a local MCP that allows LLMs to read and write to a RAG. It also has built in systems that help keep the system efficient like a max chunk size and deprecation when information gets updated.

I would love it if you all would try it out and let me know your thoughts. It's been a huge time and token saver for us.

Thanks for the help and enjoy!