Rug Pull & Scam Tokens¶
Deploy a token, add some liquidity, shill it, pull the pool, money gone. Every step is perfectly legal on-chain.
Core mechanics¶
A rug pull is "insiders exit." The standard five steps:
1. Deploy a backdoored token
├── Unlimited mint callable only by owner
├── pause (freeze all transfers)
├── blacklist (block specific addresses from selling)
└── setTax (trading tax adjustable up to 100%)
2. Add liquidity
└── Create a pool on Uniswap / PancakeSwap / Raydium to fake "depth"
3. Shill and pump
├── KOL shills, bot volume, TG hype, retail FOMO piles in
4. Pull the rug
├── Withdraw LP tokens and take the ETH/USDC out of the pool
├── Or mint a massive supply and dump the price toward zero
└── Or flip pause so nobody can sell at all
5. Vanish
└── TG group deleted, X account wiped, website offline
Variants:
| Variant | Mechanism | Victim experience |
|---|---|---|
| Honeypot | Contract allows buys but blocks transfer/sell |
Bought in, locked forever |
| Blacklist token | Owner blacklists specific addresses from selling | Only "their people" can exit |
| Hard rug | Liquidity pulled / pool assets transferred | Instant zero |
| Soft rug | Team sells slowly, KOLs dump quietly, unlock cliffs hit | Slow bleed — you're down 90% before you notice |
| Celebrity / newsjacking coin | Named after a star, politician, or breaking event | Pure sentiment; goes to zero when attention moves |
2024-2026: launchpads dropped the barrier to the floor¶
One-click token launchers (Pump.fun pioneered it on Solana, every chain followed) cut the cost of "launching a token" to tens of dollars and thirty seconds.
Consequences:
- Thousands of new tokens per day, the overwhelming majority ending at zero or rugged
- Bonding curves structurally favor early entrants — later buyers pay for the ones ahead of them
- Livestreaming, volume bots, and fake KOL matrices form a complete industry
- Individual losses are usually small (hundreds to a few thousand dollars), but the victim count is enormous and the aggregate is substantial
This isn't a security problem, it's a market structure problem
The expected return of a memecoin is negative by design. You're buying a ticket, not an asset. Treating it as gambling is fine. Treating it as investing is not.
The three-minute pre-buy check¶
1. Read the contract¶
| Check | How | Red flag |
|---|---|---|
Is there an unrestricted mint? |
Read source / GoPlus / Token Sniffer | Owner can inflate supply anytime |
Is there pause / blacklist? |
Search function names | Walk away if present |
| Can the tax change? | Look for setTax / _taxFee |
Can be set to 100% |
| Can you actually sell? | Test with a small amount | Can't sell = honeypot |
| Is it verified open-source? | Check "Verified" on the explorer | Unverified = walk |
2. Check the liquidity¶
- Is the LP locked? Unlocked means the owner can pull anytime. Check Team.Finance, UNCX,
rugcheck.xyz - Where are the LP tokens? Still with the owner, or burned / in a multisig / in a timelock?
- Depth vs market cap: a $50K-deep pool with a $5M market cap collapses on a few tens of thousands of dollars
3. Check holder distribution¶
Top 10 hold > 50% → highly concentrated, dumps anytime
One address holds > 20% → a single whale can zero this
Deployer still holds a lot → they can exit whenever
4. Check the "people"¶
- Anonymous team (normal for memecoins) — then judge community quality instead
- Vague website/roadmap, audit by a no-name firm
- TG full of bots, muted, shilling only
- KOL shills it but you can't find any position in their wallet
Scanning tools¶
| Tool | Chain | Use |
|---|---|---|
| GoPlus Token Detector | Multi-chain | Contract risk API + browser extension |
| Rugcheck.xyz | Solana | LP lock, holdings, permissions in one scan |
| Token Sniffer | Multi-chain EVM | Contract similarity + risk score |
| DEXTools | Multi-chain | Pool depth, holder distribution, live trades |
| Honeypot.is | EVM | Simulates a buy and sell to detect honeypots |
| CertiK Skynet | Multi-chain | Project security score (note: a score is not safety) |
Tools aren't insurance
Scanners only detect known patterns. A freshly written backdoor won't match. And there's now a cottage industry of "anti-scan" services helping teams write contracts that pass common scanners.
Defense checklist¶
- Test small, then actually sell once to confirm you can exit before sizing up
- Check the LP lock — no lock, no touch
- Check holder distribution — heavy top-10 concentration, walk
- Don't chase KOL shills — the person shilling is often the person preparing the rug. Their cost is zero; yours is principal
- Set a stop loss and actually honor it — zero is the default outcome for memecoins
- Size it as ticket money — only what you can lose without it mattering
- Use a burner wallet for new tokens — in case the contract is a subtler trap (malicious approvals, callback attacks)
Adjacent risks: not rugs, but they look like them¶
| Type | Distinction |
|---|---|
| Exit scam | A custodial project or exchange takes the money and disappears |
| Ponzi / high-yield scheme | Pays earlier participants with later participants' money until there are no later participants |
| Pig butchering | Long-term relationship building before the investment pitch — see Phishing & Social Engineering |
| Non-ruggable protocol, rugging team | Contracts are fine, but the team drains the treasury (which the multisig controls) |
References¶
- Rugcheck.xyz (Solana)
- GoPlus Security
- Token Sniffer
- Honeypot.is
- Rekt Leaderboard — ranked hacks and scams
- SlowMist — scam technique catalogs