Infinitum is a self-hosted information aggregation workbench designed to pre-filter personal information streams, enhancing acquisition efficiency. It targets users overwhelmed by data, offering automated RSS fetching, AI summarization, event clustering, and daily report generation for a more manageable information flow.

How It Works

The system uses a Web UI for task queuing and asynchronous Workers for execution. It ingests multi-source RSS feeds, fetching full article bodies when needed. A key differentiator is its AI pipeline: initial rule-based filtering is followed by AI analysis for summarization, translation, quality assessment, and structured event extraction. These AI-identified events are clustered to de-duplicate information before presentation in a browsable feed or AI-generated daily reports.

Quick Start & Requirements

• Docker: Copy docker-compose.yml.example to docker-compose.yml, pull images (docker compose pull), and start (docker compose up -d). Configure ADMIN_PASSWORD, ADMIN_SESSION_SECRET, SITE_URL. Access web at http://localhost:3001.
• Local Dev: Requires Node.js/npm/Prisma. Install deps (npm install), configure .env, generate schema (npm run prisma:generate), setup DB (npm run db:setup), run dev server (npm run dev) and worker (npm run worker). Local access: http://localhost:3000.
• Prerequisites: Docker, Node.js/npm. Worker CPU usage is configurable via WORKER_CPUS.

Highlighted Details

• Automated AI content summarization, translation, and structured event analysis.
• Intelligent event grouping to consolidate related news items and reduce redundancy.
• Generation of structured AI daily reports with public RSS and Markdown export options.
• Asynchronous task processing via dedicated Worker service for scalability and monitoring.
• Configurable rule-based filtering at ingestion to manage noisy sources.

Maintenance & Community

No specific details on contributors, sponsorships, or community channels were found in the provided README.

Licensing & Compatibility

Released under the MIT License, generally permitting commercial use and modification with attribution.

Limitations & Caveats

Default configurations (sources, prompts) apply only during initial setup; changes require admin interface modification. The asynchronous Worker service must be running for tasks like fetching, AI analysis, and report generation. Consult logs (app, worker) for diagnostics, especially if AI features or daily reports fail due to missing API configurations or insufficient content. Docker maps host access to http://localhost:3001.