aoogle

A meta search engine that scrapes public web indexes and re-ranks results using a multi-factor scoring algorithm. No ads, no AI summaries, no tracking.

How it works

1. Your query is sent to public web indexes
2. Results are scored across 8 signals:
   • Title match (24%) — exact phrase matches rank highest
   • Snippet relevance (18%) — term frequency and proximity
   • Domain authority (16%) — trusted sites score higher, content farms get penalized
   • Content quality (12%) — known low-quality sites (GeeksforGeeks, Guru99) hit with −20
   • Freshness (10%) — newer content gets a boost
   • Category relevance (7%) — does the result type match your intent?
   • Reddit boost (7%) — discussion queries push Reddit results up
   • URL quality (6%) — clean short URLs beat tracking-laden ones
3. Results are sorted by total score and returned — no promoted links, no sponsored slots

Features

• Clean, familiar search interface
• Image search via Bing
• Knowledge panels for popular entities
• Search suggestions (powered by Google Suggest API)
• "I'm Feeling Lucky" — random wholesome query
• Scoring is fully transparent and heuristic (no ML, no vectors, no indexing)

API

Free, tokenless search API for agents, CLI tools, and LLMs. No registration or API key needed.

GET /api/search?q=your+query

• 25 requests per hour per IP address
• Returns clean JSON with title, url, snippet, category, score, and more
• Supports pagination (&page=N) and pretty-print (&pretty=1)
• Respects crisis detection and content moderation

Documentation: /docs

Quick start:

curl "https://aoogle-production.up.railway.app/api/search?q=python+programming"

Agent helper, interactive CLI demo, and JavaScript/Node.js examples available on the docs page.

Running locally

pip install -r requirements.txt
python main.py

The server runs on http://localhost:5000.

Deployment

The app is Flask-based and ready for any WSGI server. A vercel.json is included for Vercel deployment.

Tech stack

• Python 3 + Flask
• BeautifulSoup for scraping
• DuckDuckGo HTML endpoint (web results)
• Bing Images (image results)
• Google Suggest API (autocomplete)
• In-memory caching (Redis optional)
• Chart.js (live stats dashboard)

License

Apache 2.0