Summary

KnowNote offers a local-first, open-source AI workspace designed as an alternative to Google NotebookLM. It targets learners and developers seeking privacy, control over their data, and the ability to use self-hosted or private LLM APIs without complex Docker setups. The project provides a lightweight, convenient desktop application for building intelligent knowledge bases from documents, facilitating research and writing.

How It Works

This project utilizes an Electron-based desktop application architecture. It employs Retrieval-Augmented Generation (RAG) to enable conversational querying and summarization over user-uploaded documents. KnowNote supports a provider-based design for integrating various LLM APIs, including OpenAI, DeepSeek, and Ollama. Core data storage and semantic search are handled locally using SQLite, augmented with the sqlite-vec extension for efficient vector retrieval, ensuring precise source traceability.

Quick Start & Requirements

• Installation: Download pre-built executables from the GitHub Releases page (Windows .exe, macOS .dmg). For development, clone the repository, run pnpm install, and then pnpm dev.
• Prerequisites: For development, Node.js and pnpm are required. Basic usage does not require Docker or server setup. LLM API keys are optional but necessary for AI features.
• Compatibility: Supports Windows and macOS.

Highlighted Details

• Document Support: Ingests PDF, DOCX, PPTX, and web pages, with automatic structure parsing and content extraction.
• AI Capabilities: Leverages RAG with multiple LLM provider integrations (OpenAI, DeepSeek, Ollama) for conversational querying and summarization, providing precise source attribution.
• Local-First Architecture: Emphasizes data privacy and offline usability by storing all information locally via SQLite, enhanced with vector search (sqlite-vec) for semantic retrieval.
• User Interface: Built with Electron, offering a cross-platform desktop application experience.

Maintenance & Community

The project is primarily maintained by @MrSibe and is in an early stage. Feedback, suggestions, and contributions are welcomed via GitHub Issues, Discussions, and Pull Requests, particularly concerning learning workflows, knowledge visualization, and model/provider abstraction.

Licensing & Compatibility

This project is licensed under the GPL-3.0 License. As a copyleft license, GPL-3.0 requires derivative works to also be licensed under GPL-3.0, which may impose restrictions on integration with proprietary or closed-source software.

Limitations & Caveats

The project is in its initial development phase, with some features still rough or under active development (e.g., audio processing, quiz generation). Primarily targets Windows and macOS; Linux support is not explicitly detailed. The GPL-3.0 license imposes copyleft requirements that may affect integration into proprietary or commercial projects.