Analysis updated 2026-05-18
Capture quick notes, tasks, or instructions with a keyboard shortcut from anywhere.
Ask natural language questions and get answers pulled from your own stored notes.
Keep a private, offline record of ideas and commands without a cloud account.
Track to-do items alongside general notes in one searchable place.
| erezshahaf/lore | dennis960/backapp | dexhorthy/shannon | |
|---|---|---|---|
| Stars | 208 | 209 | 204 |
| Language | TypeScript | TypeScript | TypeScript |
| Setup difficulty | moderate | moderate | moderate |
| Complexity | 3/5 | 3/5 | 3/5 |
| Audience | general | ops devops | developer |
Figures from each repo's GitHub metadata at analysis time.
Requires installing Ollama to run the local AI model that powers note capture and answers.
Lore is a desktop app that lives in your system tray and acts as a private AI-powered second brain. The idea is that you summon it with a keyboard shortcut, type anything you want to save or ask, and it handles the rest. All processing and storage happens entirely on your own machine, with no cloud services or accounts required. When you type a thought, Lore automatically figures out whether it is a note to store, a question to answer, a to-do item, or an instruction to follow going forward. It stores everything in a local vector database, which is a type of database that understands meaning rather than just exact words. When you ask a question later, it finds relevant notes and uses a local AI model to generate a useful answer based on what you have stored. Practical examples: you could save the exact command that fixed a tricky production bug, jot down notes from a meeting, add tasks to a to-do list, and then later ask "is there anything I need to do on the way home?" and Lore will search your stored notes to answer. You would use this if you want a fast, private way to capture and recall thoughts, notes, and tasks without trusting any external service with your data. It runs on Windows, macOS, and Linux. The app uses Ollama to run AI models locally and stores data in a local vector database called LanceDB. Built with TypeScript.
A private, local-only desktop second brain that captures notes and to-dos by voice or text and answers questions using a local AI model.
Mainly TypeScript. The stack also includes TypeScript, Ollama, LanceDB.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly general.
This repo across BitVibe Labs
Verify against the repo before relying on details.