sidinsearch/superbrain

SuperBrain — AI-powered save-it-later brain for Instagram, YouTube & Web

What it solves

SuperBrain is a self-hosted AI-powered content archive for Android that prevents saved content from becoming "graveyards." It solves the problem of fragmented saving across different platforms (Instagram, YouTube, web pages) where users often forget what they saved, why they saved it, and where it is located.

How it works

Users share URLs from any Android app to SuperBrain. A self-hosted Python backend then processes the content using a multi-provider AI model router. This router automatically selects the fastest available model from providers like Groq, Gemini, OpenRouter, or local Ollama to perform specific tasks:

  • Content Analysis: Generates titles, summaries, and tags for web pages and social media posts.
  • Media Processing: Transcribes audio using Whisper (cloud or local) and identifies background music in Instagram reels via Shazam.
  • ** uma YouTube Analysis**: Uses Gemini to analyze videos directly via URL without downloading them.
  • Web Scraping: Extracts text using multiple strategies (newspaper4k, trafilatura, Wayback Machine).

Everything is stored in a user-owned SQLite database, and an Android app provides the interface for searching, filtering, and organizing content into collections.

Who it’s for

Individuals who save large amounts of digital content across various platforms and want a centralized, self-hosted, and searchable AI-powered archive with no cloud subscriptions or data harvesting.

Highlights

  • Multi-provider AI Router: An EMA-ranked, self-optimizing router that automatically falls back between different AI providers to ensure maximum speed and reliability.
  • Universal Share Target: Works with any app that shares URLs, including Instagram, YouTube, and Chrome.
  • Intelligent Discovery: Full-text search across titles, summaries, and transcriptions, combined with smart "Watch Later" reminders.
  • Privacy-First: Self-hosted backend with a local SQLite database, ensuring the user owns their data.
  • Local Fallbacks: Supports local inference via Ollama and local OpenAI Whisper for transcription when cloud APIs are unavailable.

Related

  • Project
    henrydaum/second-brain

    A self-hosted personal AI assistant that indexes local files, manages emails, and executes system commands to act as a digital extension of the user's memory and machine.

  • Project
    smartscanapp/smartscan-android

    An Android app that turns images and videos into a searchable personal knowledge base using on-device processing and automatic media grouping.

  • Project
    karakeep-app/karakeep

    A self-hostable bookmarking app that uses AI for automatic tagging, summarization, and semantic search to organize large collections of links and files.

  • Project
    SilentCoderHere/aihub

    A privacy-first, open-source Android app that consolidates multiple AI assistants into a single, customizable interface with built-in ad and tracker blocking.

  • Dispatch
    OpenBrief: A Local-First Approach to Video Summarization and Knowledge Management

    Explore OpenBrief, an open-source desktop application that transforms video and audio into grounded summaries and searchable briefings using local-first AI.