reorproject/reor

Health v2 · maintenance only

Flags: archived

Adoption not part of the score

8568 stars · 531 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

Reor is a local-first, AI-powered desktop note-taking app that automatically links related notes, answers questions over your notes via RAG, and provides semantic search. All notes are stored as markdown locally, with LLMs and embedding models running on-device through Ollama, Transformers.js, and LanceDB.

Use cases

• take private notes with automatic AI linking between related ideas
• ask questions about my own notes with a local LLM
• semantically search my personal knowledge base without the cloud
• run a second brain app that works fully offline
• import markdown notes and get AI-powered connections
• chat with my notes using Ollama models locally

When to choose

• you want a private, local-first alternative to cloud AI note apps like Notion AI
• you already use markdown/Obsidian-style notes and want built-in RAG and semantic search
• you can run local models via Ollama on your machine

When to avoid

• you need cloud sync and multi-device collaboration out of the box
• your hardware is too weak to run local LLMs and embedding models
• you need a web-based or mobile note-taking app

Facets

application · maturity active

rag vector-database llm-inference search-engine markdown editor developer-tools large-language-models self-hosted windows note-taking pkm local-first second-brain ollama semantic-search electron-app privacy natural-language-processing retrieval-augmented-generation macos linux desktop

Member repositories

For agents

markdown · JSON · MCP: product_card(name="reorproject/reor")