notactuallytreyanastasio
MCP Servernotactuallytreyanastasiopublic

roasted

一个基于浏览器历史记录模式幽默调侃用户的工具,支持多浏览器和隐私保护。

Repository Info

1
Stars
0
Forks
1
Watchers
0
Issues
TypeScript
Language
-
License

About This Server

一个基于浏览器历史记录模式幽默调侃用户的工具,支持多浏览器和隐私保护。

Model Context Protocol (MCP) - This server can be integrated with AI applications to provide additional context and capabilities, enabling enhanced AI interactions and functionality.

Documentation

🔥 Roasted MCP

Screenshot 2025-05-30 at 10 47 30 AM

An MCP (Model Context Protocol) server that hilariously roasts you based on your browser history patterns. Because someone needs to call out your 3 AM Wikipedia rabbit holes and your questionable shopping habits.

⚠️ Common Issues:

  • SQLITE_CANTOPEN errors: Permissions issue. macOS: Grant Full Disk Access to Claude Desktop. Windows: Run as Administrator.
  • "Could not attach to MCP server": Claude can't connect despite server running. Check config file path and restart Claude Desktop completely (⌘+Q, not just close window).

See TROUBLESHOOTING.md for detailed fixes.

🎯 Features

  • Multi-Browser Support: Extracts history from Chrome, Edge, Brave, Opera, Vivaldi (and Safari on macOS)
  • Intelligent Pattern Analysis: Identifies procrastination habits, social media addiction, shopping sprees, and weird browsing hours
  • Customizable Roast Intensity: Choose from gentle ribbing to savage burns
  • Privacy-First Design: Built-in filtering for sensitive content (health, finance, dating sites)
  • Organic Browsing Detection: Distinguishes between natural browsing and intentional searches

🚀 Quick Start for Developers

⚡ Super Quick Setup (5 minutes)

Requirements: Node.js 18+, Windows/macOS/Linux, Claude Desktop

git clone https://github.com/notactuallytreyanastasio/roasted.git
cd roasted

# Auto-detect your platform and run setup
node setup-cross-platform.js

# OR run platform-specific setup:
# Windows: setup.bat
# macOS/Linux: ./setup.sh

That's it! The setup script handles everything automatically.

🧪 See It In Action First

Want to see what you're in for? Run the demo:

node demo.js

This shows example outputs and demonstrates all roast severity levels with your actual browser data.

📚 Detailed Documentation

  • Developer Setup Guide - Step-by-step setup with troubleshooting
  • Troubleshooting Guide - Fix any issues you encounter
  • Example Outputs - See what the roasts look like
  • Main Documentation - Continue reading below for full details

✅ Verify Everything Works

After setup, run a quick health check:

./verify.sh

This verifies all components are working correctly.

🎯 One-Liner Test

Test the MCP server directly:

echo '{"jsonrpc": "2.0", "id": 1, "method": "tools/call", "params": {"name": "roast_browser_history", "arguments": {"severity": "gentle"}}}' | node dist/index.js

First Roast in Claude

Once configured, ask Claude:

Can you roast my browser history from the last 7 days with medium severity?

🛠️ Available Tools

roast_browser_history

Generates a humorous roast based on your browsing patterns.

Parameters:

  • days (number, default: 7): Number of days to analyze
  • severity (string, default: "medium"): Roast intensity ("gentle", "medium", "savage")
  • include_chrome (boolean, default: true): Include Chrome-based browsers
  • include_safari (boolean, default: true): Include Safari
  • privacy_level (string, default: "default"): Privacy filtering ("default", "paranoid", "custom")
  • exclude_sensitive (boolean, default: true): Filter out health, finance, dating sites
  • exclude_work (boolean, default: false): Filter out work-related sites

analyze_browsing_patterns

Analyzes browsing patterns without generating roasts - for the data nerds.

Parameters:

  • days (number, default: 7): Number of days to analyze

🔒 Privacy & Security

Your privacy is important (even when we're roasting you):

  • Local Processing: All analysis happens locally on your machine
  • No Data Transmission: Browser history never leaves your computer
  • Sensitive Site Filtering: Automatically excludes health, finance, and dating sites by default
  • URL Sanitization: Removes query parameters and personal information from URLs
  • Title Cleaning: Strips potential personal data (emails, phone numbers, SSNs) from page titles

Privacy Levels

  • Default: Excludes clearly sensitive sites but allows most content
  • Paranoid: Excludes sensitive sites, work sites, personal sites, and night browsing
  • Custom: Configure exactly what to include/exclude

🎭 Roast Severity Levels

Gentle 😊

Light-hearted observations about your browsing habits. Perfect for sharing with friends.

Medium 😏 (Default)

Sarcastic commentary with a good balance of humor and mild embarrassment.

Savage 🔥

No mercy. Brutal honesty about your digital lifestyle choices. Use at your own risk.

📊 What Gets Analyzed

The roasting algorithm looks for:

  • Social Media Addiction: Excessive Facebook, Instagram, Twitter, TikTok usage
  • Procrastination Patterns: YouTube binges, Reddit spirals, endless scrolling
  • Shopping Habits: Amazon addiction, impulse buying patterns
  • Time Patterns: Late night browsing, work hour distractions
  • Domain Dominance: Sites you visit way too often
  • Productivity Score: How much of your browsing is actually useful

🚫 What We DON'T Roast

To keep things fun (not harmful), we automatically exclude:

  • Health-related sites: Medical conditions, symptoms, treatments
  • Financial sites: Banking, investment, tax preparation
  • Dating sites: Because love is complicated enough
  • Legal sites: Your legal troubles are your own
  • Mental health resources: We support your wellness journey
  • Adult content: Keeping it PG(-13)

🔧 Development

Project Structure

src/
├── index.ts              # Main MCP server
├── browser-history.ts    # Browser history extraction
├── roast-generator.ts    # Roasting logic and templates
└── privacy-filter.ts     # Privacy controls and content filtering

Build Commands

npm run build      # Compile TypeScript
npm run dev        # Watch mode for development
npm test          # Run tests (coming soon™)

Testing Locally

node test-local.js  # Quick test of browser history extraction

🤝 Contributing

Got better roast material? Found a bug? Want to add support for more browsers?

  1. Fork the repository
  2. Create a feature branch: git checkout -b feature/better-burns
  3. Make your changes
  4. Run tests: npm test (when they exist)
  5. Submit a pull request

⚠️ Disclaimers

  • For Entertainment Only: This tool is meant for fun, not psychological analysis
  • Cross-Platform: Works on Windows, macOS, and Linux
  • Browser Permissions: May require additional system permissions to access browser databases
  • Your Feelings: We are not responsible for any emotional damage caused by accurate roasts
  • Productivity: This tool may cause you to reflect on your digital habits (sorry)

🙋‍♀️ FAQ

Q: Will this work on Windows/Linux? A: Yes! Now supports Windows, macOS, and Linux with automatic platform detection.

Q: Can I roast my friend's browser history? A: Only if they run it on their machine. We don't support remote history access (that would be creepy).

Q: The roast wasn't funny enough. Can I get a refund? A: Try the "savage" setting. Also, humor is subjective, but your browsing habits probably aren't.

Q: It says I need Full Disk Access. Is this safe? A: The tool only reads browser history databases. Check the source code - it's all open source!

Q: Why is my productivity score so low? A: That's between you and your browser history. We just report the facts.

📜 License

MIT License - Feel free to roast responsibly.


Remember: The best roasts come from a place of love. We're all just trying to survive the internet together. 🌐❤️

Quick Start

1

Clone the repository

git clone https://github.com/notactuallytreyanastasio/roasted
2

Install dependencies

cd roasted
npm install
3

Follow the documentation

Check the repository's README.md file for specific installation and usage instructions.

Repository Details

Ownernotactuallytreyanastasio
Reporoasted
LanguageTypeScript
License-
Last fetched8/10/2025

Recommended MCP Servers

💬

Discord MCP

Enable AI assistants to seamlessly interact with Discord servers, channels, and messages.

integrationsdiscordchat
🔗

Knit MCP

Connect AI agents to 200+ SaaS applications and automate workflows.

integrationsautomationsaas
🕷️

Apify MCP Server

Deploy and interact with Apify actors for web scraping and data extraction.

apifycrawlerdata
🌐

BrowserStack MCP

BrowserStack MCP Server for automated testing across multiple browsers.

testingqabrowsers

Zapier MCP

A Zapier server that provides automation capabilities for various apps.

zapierautomation