Self-hostable bookmark app with AI tagging, OCR, and full text search.
Self-hostable bookmark-everything app with AI-based automatic tagging and full text search.
Karakeep is a self-hostable app designed for capturing links, notes, images, and PDFs with a touch of AI. It is ideal for data hoarders and professionals who want to curate and organize unstructured web content with automated intelligence.
The application automatically fetches link metadata, runs OCR on stored images, and extracts text to protect against link rot using monolith and yt-dlp. It features LLM-based automatic tagging, summarization, semantic search via Meilisearch, and an agent-friendly CLI.
It suits self-hosters and digital researchers who want total control over their saved content and read-it-later workflows. It does not suit users looking for a fully managed, zero-setup commercial service without touching Docker or a home server.
Automatically tags and summarizes your saved content with support for local models via Ollama.
Protects against link rot using monolith for full page archiving and yt-dlp for video archiving.
Extracts text directly from stored images to make all your hoarded graphics searchable.
Includes a powerful command line interface and official skills designed to be friendly with LLM agents.
Free under the AGPL-3.0 licence if you self-host it on your own hardware. A managed cloud subscription is also available for users who prefer not to handle their own server infrastructure.
880 open remote AI-training roles, updated November 15, 2026; most pay $20–$175 an hour. Zekai lists them free; the application goes to the employer’s platform.
Yes, Karakeep supports local AI models for automatic tagging and summarization using Ollama.
It uses monolith to create full page archives and yt-dlp for automatic video archiving.
Yes, a managed Karakeep Cloud option is available at cloud.karakeep.app to support project development.
See Zekai first in Google