BrowseMemory

About

Your browser sees everything you read. It remembers almost nothing useful.

BrowseMemory fixes that. Every page you visit gets automatically 
indexed using on-device AI — no setup, no API key, no account. 
Later, search what you read using natural language. Even if you 
don't remember the title, the URL, or where you found it.

Type "that article about React performance I read last week" and 
find it instantly. Search by concept, not by keyword.

──────────────────────────────────────
🔍 SEMANTIC SEARCH
──────────────────────────────────────
Search your entire reading history by meaning — not exact words.
"frontend rendering optimization" finds "Why Your React App 
Renders Too Much" even with zero shared keywords.

──────────────────────────────────────
🤖 100% ON-DEVICE AI
──────────────────────────────────────
All AI processing runs locally in your browser using WebAssembly.
No API key required. No cloud calls. Works completely offline 
after the first model load.

──────────────────────────────────────
🔒 GENUINELY PRIVATE
──────────────────────────────────────
Your browsing history is among the most sensitive data you 
generate. BrowseMemory stores everything — pages, summaries, 
and embeddings — exclusively on your device.

- No account required
- No sign-up, no email
- No servers — we literally cannot see your data
- No data ever leaves your browser

──────────────────────────────────────
⚡ AUTOMATIC — ZERO EFFORT
──────────────────────────────────────
Visit pages normally. BrowseMemory indexes them silently in the 
background. No clicking Save, no manual tagging, no bookmarking. 
Your reading history builds itself.

──────────────────────────────────────
🛡️ YOU ARE IN CONTROL
──────────────────────────────────────
- Block any site from being indexed
- Clear your entire index anytime
- See exactly how many pages are stored
- Your data belongs to you

──────────────────────────────────────
HOW IT WORKS
──────────────────────────────────────
1. Install BrowseMemory
2. Browse normally — no setup needed
3. Click the extension icon → Open Search
4. Type what you remember reading about
5. Find it instantly

──────────────────────────────────────
PERFECT FOR
──────────────────────────────────────
- Developers researching across multiple sessions
- Researchers managing large volumes of reference material  
- Knowledge workers who read extensively but lose track of sources
- Anyone tired of re-Googling things they've already read
- Privacy-conscious users unwilling to share browsing history 
  with cloud services

──────────────────────────────────────
TECHNICAL NOTES
──────────────────────────────────────
BrowseMemory uses the all-MiniLM-L6-v2 embedding model (22MB) 
running via Transformers.js and ONNX Runtime WebAssembly. The 
model downloads once on first install and is cached locally. 
All search uses cosine vector similarity — no string matching.

Note: Clearing browser data (Cookies and site data) will also 
clear your BrowseMemory index. Export functionality coming soon.