Magnemo-AI/magnemo

Overview

Magnemo is a governed memory system for AI agents, implemented as an MCP server. It stores memories locally in plain files and issues a receipt for every entry, providing an auditable trail. Written in Python, it is designed for developers who want a simple, local memory layer for their agents.

Best for

Best for
Developers needing a lightweight, auditable memory layer for AI agents.

Use cases

• Store and retrieve agent memories locally
• Track memory entries with receipts for audit
• Integrate memory into MCP-compatible agent workflows

How to use

Install

pip install magnemo

Tested with

Claude Code, Cursor, Windsurf

Notes

Magnemo is a governed memory system for AI agents, implemented as an MCP server. It stores memories locally in plain files and issues a receipt for every entry, providing an auditable trail. Written in Python, it is designed for developers who want a simple, local memory layer for their agents.

2 stars on GitHub. Last updated 2026-09-17. Licensed Apache-2.0.

Use cases

• Store and retrieve agent memories locally
• Track memory entries with receipts for audit
• Integrate memory into MCP-compatible agent workflows

Pros

• Local plain files are simple and transparent
• Receipt on every entry enables auditability
• Python-based, easy to extend

Cons

• Very low adoption (2 stars) suggests limited maturity
• No evidence of advanced features like search or indexing
• Requires MCP server setup, adds complexity

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.