Skip to content

gyoomei/mimorug

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 

Repository files navigation

🛡️ MiMoRug

Rugpull Risk Predictor — Powered by Xiaomi MiMo V2.5

Paste any Ethereum token contract. Get a 0–100 rug probability score with 12-signal forensic breakdown — explained in plain English by Xiaomi MiMo V2.5.

Live Demo Try Now AI Data License


📖 The Problem

Every week, thousands of new tokens deploy on Ethereum. Most are scams. Existing tools like TokenSniffer flag known patterns but don't reason about combined signals — and most charge for premium features.

MiMoRug closes the gap. Paste a contract, get an actionable verdict.

The audit engine pulls 40+ security fields from GoPlus, scores 12 weighted risk signals, and lets MiMo V2.5 compose a plain-English verdict. Result: a 0–100 rug score backed by traceable reasoning, free for everyone.


✨ Core Features

🎯 12 Risk Signals

Signal Weight What It Catches
Honeypot 30 Cannot sell after buying — pure scam
Mintable Supply 20 Owner can dilute holders infinitely
Hidden Owner 20 Control obscured — likely rug prep
Sell Tax >20% 20 Extraction tax disguised as fee
Take-back Ownership 15 Renounce reversible — fake decentralization
Blacklist Function 15 Owner can freeze any wallet
Closed Source 15 Can't audit — black box
Holder Concentration 15 Top 10 wallets dominate supply
Trading Pausable 10 Owner can halt trading anytime
Buy/Sell Tax 5–10% 10 Moderate extraction risk
LP Risk 10 Few/unlocked LPs — soft rug enabler
DEX Listing 5 No liquidity available
Self-destruct +25 Contract can erase itself
Upgradeable Proxy +8 Logic can change without warning

📊 Verdict Tiers

Score Tier Action
0–14 ✅ LOOKS SAFE Standard DeFi caution applies
15–39 🟡 PROCEED WITH CAUTION Verify each warning
40–69 🟠 HIGH RISK Avoid unless experienced
70–100 🚨 RUG ALERT Do NOT interact

🌐 Other

  • 🌍 Bilingual EN / ID full translation
  • 🌓 Dark / Light theme (WCAG-AA contrast)
  • 📱 Mobile-first responsive (clamp 320px → 1440px)
  • 🔗 Shareable URL (hash-based deep link)
  • 🚀 Single zero-dependency HTML on GitHub Pages

🏗️ Architecture

INPUT: Token contract (0x...)
    ↓
GoPlus Security API → 40-field audit (no API key)
    ↓
12-Signal Risk Analyzer → weighted score 0-100
    ↓
MiMo V2.5 Narrative Engine → plain-English verdict
    ↓
SVG Risk Gauge + Signal Grid + Stats

Three free APIs power the audit:

  • GoPlus Security — honeypot/owner/tax/holder analysis
  • Blockscout — contract verification, creation tx history
  • CoinGecko — market data when available

🚀 Quick Start

git clone https://github.com/gyoomei/mimorug.git
cd mimorug
python3 -m http.server 8080
# Open http://localhost:8080

Or just visit the live demo.


📁 Project Structure

mimorug/
├── index.html       # UI + CSS (15 KB)
├── app.js           # Audit engine + i18n + render (28 KB)
└── README.md

Total: ~45 KB. Zero npm dependencies. Loads in <1s.


🎨 Risk Tier Examples

Token Score Verdict
USDC ~18 🟡 CAUTION (proxy upgradeable)
PEPE ~15 🟡 CAUTION (memecoin standards)
SHIB ~12 🟡 CAUTION (older contract)
Honeypot scam 70+ 🚨 RUG ALERT

🛣️ Roadmap

  • 12-signal weighted scoring
  • GoPlus integration (40 fields)
  • MiMo V2.5 plain-English verdict
  • SVG risk gauge
  • Bilingual EN / ID
  • Dark / Light theme
  • Mobile responsive
  • Multi-chain (Base, Arbitrum, BSC)
  • Dev wallet history scoring (multi-token rug pattern)
  • Comparison against known rug DB
  • Telegram bot integration

📸 Screenshots

See screenshots in this repo or live demo.


📜 License

MIT © 2026 — built with 🧠 by Xiaomi MiMo V2.5


🔗 Links


Built for users who got rugged once and never want to again.

About

Multi-chain rugpull risk scanner • Audit any token contract across 10 EVM networks via 12-signal weighted analysis covering honeypots, hidden owners, holder concentration, and liquidity locks • 0–100 rug probability with plain-English red flag breakdown • Powered by Xiaomi MiMo V2.5 • Live data • No API key • Single HTML

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors