Quant Bot
Not Just Alerts.
Intelligence.
Most scanners blast every new token. QuantRugGuard runs 155+ candidates through 7 filter layers every 10 seconds — and only alerts when something actually deserves your attention.
- 7 Filter Layers
- 10s Scan Interval
- 155+ Candidates Per Cycle
- 68% 2x Win Rate
99% of Tokens
Never Make It.
Hundreds of new tokens launch on Solana every single hour. The overwhelming majority never complete their bonding curve — they die quietly on pump.fun, drained by bots and insiders before a single real trader gets a chance. Of those that do graduate to Raydium, most rug within minutes. The rest bleed out slowly as early wallets rotate out. The token you buy matters less than when you hear about it and who filtered it before it reached you.
QuantRugGuard scans over 155 candidates every 10 seconds. In a typical hour, the vast majority get rejected — wrong mcap, no LP lock, bundled launch, insider concentration, weak buy pressure. What comes through the other side is the top 1% that actually shows the ingredients of a real move.
Timing Is
Everything.
The difference between a 5x and a missed trade is often minutes — sometimes seconds. QuantRugGuard fires T1A alerts when a token is still at $8k–$20k market cap, immediately after graduating from pump.fun to Raydium. At that size, a relatively small amount of buying moves the price dramatically. By the time most people see it trending on DexScreener, the move is already done.
Speed without filtering is just noise. That's why the rug filter runs first — before the momentum check, before the signal score, before anything reaches you. You're not just getting fast alerts. You're getting fast alerts on tokens that have already cleared a bar that most scanners don't even attempt to set.
Information Others
Don't Have.
Most traders rely on the same public DexScreener feed everyone else watches. QuantRugGuard pulls from 6 parallel sources simultaneously — including Helius webhooks that fire the moment a token event hits the chain. When a developer wallet sells, you know before it shows up on any chart. When a token's bonding curve completes and it graduates to Raydium, you know within seconds.
That's a structural edge. Not a signal tip, not a call from a KOL who already entered — actual infrastructure that processes information faster and more thoroughly than doing it manually. That's what a 68% win rate on 2x+ looks like in practice.
Rugs Don't Make
It Through.
The biggest cost in memecoin trading isn't missing a pump — it's the rugs. One bad entry can wipe out multiple wins. QuantRugGuard's hard gates are non-negotiable: mint authority unrevoked, instant permanent block. Freeze authority unrevoked, instant permanent block. RugCheck marks the deployer as a serial rugger, instant permanent block. These checks don't get loosened, they don't have exceptions, and they run on every single candidate regardless of how good everything else looks.
The soft gates cover the subtler risks — high insider concentration, bundled launch manipulation, developer holdings, unlocked liquidity. Combined, they eliminate the category of trades that feel fine until they aren't. What's left is a tighter, cleaner set of opportunities — which is exactly what a consistently positive win rate requires.
Every Other Scanner
Is Noise.
The Solana memecoin space moves fast. Most tools can't keep up — and the ones that do alert on everything, leaving you to do the filtering yourself.
❌ The Old Way
- Alerts fire on every new token — 100+ per hour
- No rug detection — you find out after you're in
- No momentum gating — plenty of dead tokens with volume
- Static filters — can't adapt to market conditions
- No follow-up — you don't know if it pumped or rugged
- No dev sell detection — biggest red flag, missed entirely
- Bundler bots sneak through — coordinated launches look organic
✅ QuantRugGuard
- Tight filters mean 1–5 quality alerts per hour, not 100
- 7-layer rug filter runs before every single alert
- USD buy/sell ratio gates — real buying pressure only
- AI signal score adapts based on win/loss outcomes over time
- Milestone tracking fires at 2x, 3x, 5x, 10x — automatically
- Dev sell webhook via Helius fires the moment it happens
- Bundle detection blocks coordinated launch manipulation
10 Seconds.
Every Cycle.
Most scanners poll DexScreener every 30–60 seconds. By the time they alert, the entry is gone. QuantRugGuard runs a full scan cycle every 10 seconds — 6 source feeds in parallel, 155+ candidates evaluated, filters applied, alert fired.
Parallel Source Fetching
New pairs, DexScreener profiles, token boosts, top boosts, Helius webhooks, and search — all 6 sources run simultaneously via ThreadPoolExecutor with an 8-second hard timeout.
Helius WebSocket
A dedicated Helius WebSocket service runs alongside the scanner — new token events and dev wallet transactions arrive in real-time, not on a poll schedule.
25-Second Candidate Cache
The candidate pool caches for 25 seconds to prevent DexScreener rate limiting during heavy scan periods. Fresh enough to catch moves, smart enough not to get throttled.
7 Layers Between
You and a Rug.
Every candidate passes through every layer in order. Fail any hard gate — instant permanent rejection. Fail a soft gate — blocked for your session. Only tokens that clear all 7 ever reach you.
Mint & Freeze Authority
Checks via Helius RPC whether the token's mint or freeze authority is still active. If either is unrevoked, the dev can create infinite supply or freeze your wallet — permanent hard reject, never alerts again.
Rugged / Serial Rugger Detection
RugCheck flags whether a token has already been marked rugged, or if the deployer wallet has a history of previous rugs. Either flag = permanent hard reject across all future scans.
Liquidity Pool Lock
Verified via RugCheck first, then Helius directly for tokens too new to be indexed. Unlocked LP means the dev can pull the pool at any time. PumpSwap tokens pass automatically — LP lock is enforced by the protocol.
Bundle Detection
Checks what % of the token's supply was purchased in coordinated bundle transactions at launch. Default maximum: 5%. Bundled launches are almost always coordinated manipulation — sniper bots buying before anyone else.
Top-10 Wallet Concentration
The top 10 holder wallets (excluding LP) must hold under 35% combined. High concentration means a single whale can crater the price. LP wallets are filtered out so the gate measures real holder concentration, not pool depth.
Dev Wallet Holdings
Developer wallet must hold under 5% of supply. High dev holdings are a direct rug vector — the moment they sell, you're holding the bag. Combined with the serial rugger check, this covers both current and historical dev risk.
Momentum & Liquidity Gates
Buy/sell USD volume ratio (min 1.2x), vol/mcap ratio, minimum liquidity ($15k+), minimum makers, and age window. These confirm real organic buying pressure — not wash trading or a single large wallet moving the numbers.
/myset.Two Scores.
One Decision.
Every alert that passes the rug filter gets two independent scores — a Signal Score measuring momentum quality, and a Rug Score measuring residual risk. Both appear on every alert.
What Goes Into the Signal Score
Buy/Sell Volume Ratio
USD buy volume divided by USD sell volume over the last 5 minutes. Measures real demand — not transaction count, not price, actual dollar flow direction.
Vol / Market Cap Ratio
Volume as a percentage of market cap. A token doing 25%+ of its mcap in 5-minute volume is showing genuine interest relative to its size.
Holder Velocity
How fast unique holders are accumulating. Organic growth looks different from a single whale — the score weighs maker count and holder spread together.
Adaptive Thresholds
Signal thresholds auto-tune based on win/loss outcomes over time. Filters that produce false positives get tightened. Filters that block winners get loosened. The scanner learns.
Built on the Best
Solana Infrastructure.
Three best-in-class data providers, each doing what they're best at. No single point of failure — if one source is slow, the others cover.
The fastest Solana RPC provider. Used for anything that needs to happen in real-time — token authority checks, LP lock verification for new tokens, dev wallet monitoring via webhooks.
- Mint & freeze authority checks via RPC
- LP lock verification for tokens too new for RugCheck
- Dev sell webhooks — fires the moment a wallet moves tokens
- WebSocket connection for live token events
The primary source for all market data — pair prices, buy/sell ratios, liquidity depth, volume, holder counts, and the raw candidate pool of new Solana tokens.
- 155+ candidate pairs sourced every scan cycle
- Real-time price, mcap, liquidity, volume per pair
- Buy/sell transaction counts and USD volume
- New pair detection — catches tokens within minutes of launch
Deep security analysis that goes beyond surface-level checks — cluster detection, insider wallet identification, historical deployer reputation, and a composite rug score per token.
- Composite rug score (0–100) per token
- Bundle & coordinated wallet cluster detection
- Top-10 holder concentration (LP excluded)
- Dev wallet history — serial rugger detection
- Pre-fetched background cache — data ready before it's needed
How It Stacks Up.
Most Telegram scanners are wrappers around a single DexScreener endpoint with a price threshold. Here's the difference.
| Feature | QuantRugGuard | Generic Scanners | Manual Watching |
|---|---|---|---|
| Scan interval | 10 seconds | 30–60 seconds | Whenever you check |
| Candidates per cycle | 155+ | 20–50 | 1 token at a time |
| Mint authority check | ✓ Hard gate | ✗ Not checked | Manual lookup |
| LP lock verification | ✓ Helius + RugCheck | ✗ Not checked | Manual lookup |
| Bundle detection | ✓ Max 5% | ✗ Not checked | ✗ Not checked |
| Dev wallet tracking | ✓ Real-time webhook | ✗ Not tracked | ✗ Not tracked |
| Buy/sell USD ratio gate | ✓ 1.2x minimum | Transaction count only | Manual check |
| Signal score | ✓ 0–100 adaptive | ✗ No scoring | ✗ No scoring |
| Milestone tracking | ✓ 2x–100x auto | ✗ Single alert only | Manual watching |
| Personal filters | ✓ 17 adjustable | Basic thresholds | N/A |
| Win rate tracking | ✓ Per-batch stats | ✗ No tracking | Manual journal |
| Adaptive thresholds | ✓ Learns from outcomes | ✗ Static | ✗ N/A |
The Numbers
Don't Lie.
Tracked automatically since launch. Every alert logged, every outcome recorded — wins, losses, and everything in between.
- 68% 2x+ Win Rate
- Resolved Alerts
- 400+ Tokens Alerted and Counting
- 10x Best Single Alert
- $Mythos · 8h 15m
- <22m Avg. Time to 2x T1 alerts
Recent Notable Alerts
- 10x$MythosT2 · 8h 15m
- 5.3x$PropellerT1 · 14m
- 5.1x$HONKT2 · 18m
- 3.4x$ELGATOT1 · 22m
- 3x$NolanT1 · 35m
- 2.7x$WhiteHouseAppT2 · 28m
Results shown with 24h delay. Past performance does not guarantee future results. DYOR.