Reor

Stale
GitHub JavaScript AGPL-3.0

Description

A private and local AI personal knowledge management app. All data and processing stay on-device with built-in RAG, semantic search, and knowledge graph features for managing personal knowledge bases with full privacy.

Key Features

  • Local-first AI knowledge management: all data and model processing runs on-device for full privacy
  • Automatic note linking: connects related notes into a knowledge graph based on vector similarity
  • RAG Q&A mode: performs retrieval-augmented generation over your note corpus to answer questions
  • Semantic search: searches across all notes at the semantic level, not just keyword matching
  • Obsidian-style editor: supports Markdown editing with Obsidian-compatible note formats
  • Local LLM integration: downloads and runs local models directly via Ollama within the app

Use Cases

💡 Personal knowledge base management: automatically organizing scattered notes into a connected network
💡 Academic research assistant: quickly retrieving related literature notes and generating summary answers
💡 Writing and creativity tool: discovering hidden connections between notes through semantic linking for inspiration
💡 Journal and life logging: performing semantic queries and knowledge extraction on long-term accumulated journals

Tags

rag javascript desktop vector-database memory embedding

Categories

📚 RAG Tools 🧩 Agent Memory

Quick Start

1. Download the desktop app from reorproject.org (Mac/Linux/Windows)
2. Select a note storage directory on first launch
3. Start creating Markdown notes
4. Configure local models (Ollama) or OpenAI API in Settings
5. Use the sidebar to view related notes, or switch to Q&A mode to ask questions
Visit GitHub

Related Projects