LuumynOS Neural Web Search Engine

LuumynOS Neural Web Search Engine is an independent, community‑driven index engine. It crawls and archives websites submitted by the community, presenting them through a clean, space‑themed interface. Every entry is reviewed for quality and safety before joining the archive.

Created by Timothy as part of the LuumynOS project - a ceremonial operating‑system philosophy applied to how we explore, preserve, and navigate the web.

• ✓Safe search - harmful, adult, violent, hateful, and malicious content is blocked at submission, crawler, and review stages.
• ✓Community‑driven - anyone may submit a site for ceremonial indexing.
• ✓Full‑text search across titles, descriptions, domains, and metadata.
• ✓Duplicate detection via MD5 content hashing.
• ✓UTM attribution for outbound link analytics.
• ✓Privacy‑respecting - no ads, tracking, or profiling.
• ✓Admin console - real‑time submission review and index management.

LuumynOS Neural Web Search Engine does not index adult, pornographic, violent, hateful, or harmful content. Spam, phishing, and malware sites are also prohibited.

All submissions require manual admin approval before crawling begins.

The LuumynOS Neural Web Search Engine runs on a Python crawler node with a thread pool, domain rate limits, and a balanced queue system. Built for long runs and low RAM usage.

• ✓Worker Threads: 20 workers running in parallel.
• ✓Queue System: Domain-balanced FIFO queue, auto‑prune, retry, canonical checks.
• ✓Memory Use: ~1.7GB RAM during long sessions (100+ days tested).
• ✓Content Hashing: MD5 dedupe so identical pages only index once.
• ✓Safe Filter: Blocks adult, violent, hateful, harmful content at all stages.
• ✓Heartbeat: Live worker status, queue depth, uptime, crawl rate.
• ✓GUI: Tkinter dashboard with logs, worker activity, RAM, starfield.
• ✓Index: JSONL archive, portable, no database needed.
• ✓Frontend: PHP streaming results directly from index.
• ✓Admin: Review submissions, manage seeds, inspect index.

Built for stability, low resource use, and ceremonial web exploration, each part runs as a LuumynOS subsystem.

LuumynOS treats the web as a living archive - signals worth preserving, the engine is ceremonial by design, intentional, human‑curated, and not driven by commercial algorithms.

It exists to explore, not to monetise.

Know a site that deserves to be in the archive? Submit it here - ALL site submissions are reviewed before crawling.