README.md

---
title: Logica Mind — Memory Playground
emoji: 🧠
colorFrom: indigo
colorTo: green
sdk: gradio
sdk_version: 5.9.1
python_version: "3.11"
app_file: app.py
pinned: true
license: apache-2.0
short_description: Watch the memory engine extract facts from any language
tags:
  - agent-memory
  - knowledge-graph
  - llm
---

# 🧠 Logica Mind — Memory Playground

Paste any message and watch [Logica Mind](https://github.com/Rovemark/logica-mind)'s
memory engine **think**: detect the language, extract atomic facts, and tag each
with an open category + a life/work **dimension** (a 34-dimension taxonomy).

Works **offline** by default (heuristic). Paste your own OpenAI key to see the
full LLM extraction (one message → several atomic facts) — the key is used only
for that call and never stored.

Logica Mind is open-source (Apache-2.0) long-term memory for AI agents: episodic +
semantic memory, a temporal knowledge graph, an MCP server, and a self-hosted
dashboard. On LoCoMo it scores **72.5%** — above Mem0 and Zep — at ~35× fewer LLM
calls. `pip install logica-mind`