Anatomy of a rug pull: how to spot one before it happens
Updated 2 September 2026 · Belphor Research · the data behind this guide
A rug pull is not bad luck. It is a mechanism, built into a token before you ever bought it, and most of its parts are visible on-chain to anyone who checks. This guide walks through how rugs actually work on Solana, and the specific checks that catch the majority of them in advance.
- Most rugs are prepared before launch: unrevoked mint authority, hidden supply concentration, or a pool built to be drained are all readable on-chain.
- The five main rug mechanisms are liquidity pulls, mint inflation, freeze traps, honeypots (coins you can buy but not sell), and slow rugs by insider distribution.
- A sell-route test is the single strongest honeypot check: if a swap router cannot quote a route back to SOL, you are looking at a one-way market.
- No check is perfect. Rug defense is stacking filters so that each class of rug has at least one tripwire in front of it.
What a rug pull actually is
A rug pull is any design where the people who control a token can take the value out of it faster than holders can react. The name comes from the oldest version, pulling the liquidity out from under the market, but the family is bigger than that. What unites every rug is preparation: the exit was built before the entrance, and the promotion you saw was the mechanism's delivery system.
That preparation is also the weakness. Because the exit has to be built into the token or the pool, it usually leaves fingerprints on the blockchain that are visible before the rug fires.
The five mechanisms
Liquidity pull. The operator holds the LP tokens for the trading pool and withdraws both sides of it, instantly converting everyone's coins into something unsellable. Defense: check whether LP tokens are burned or locked, and treat sudden liquidity drains as an exit signal in their own right.
Mint inflation. If the token's mint authority was never revoked, whoever holds it can print unlimited new supply and sell it into the pool. The price does not have to go to zero for you to lose; dilution does the work. Defense: refuse tokens whose mint authority is still active. This is a single field on the mint account, readable in one call.
Freeze trap. An active freeze authority lets the operator freeze individual holders' token accounts, so you keep your coins but can never move them, while insiders sell. Defense: same as above, the authority is either revoked or it is not.
Honeypot. The token's transfer logic or pool is constructed so buying works and selling fails, or selling carries a tax so large it amounts to confiscation. Defense: test the exit before entering. Ask a router to quote the reverse swap; no route back to SOL, or a round-trip loss far beyond normal spread, means the market is a lobster trap.
Slow rug. Nothing dramatic ever happens. Insiders holding a large share of supply simply distribute into every rally until nothing is left. Defense: holder concentration. When a handful of connected wallets hold most of the float, you are their exit liquidity by definition.
The pre-entry checklist
Compressed into one list: is the mint authority revoked, is the freeze authority revoked, is the LP burned or locked, does a sell route exist right now, what would a round trip cost, how concentrated are the top holders, and does the deployer's wallet history contain previous rugs. Every item is on-chain. None requires trusting a chart, a website, or a Telegram admin.
One more that most people miss: ticker recycling. Ruggers reuse the names of coins that just died, because the search traffic is free. A ticker that rugged recently and reappeared under a new mint deserves automatic suspicion.
Why automation matters here
None of these checks is hard. The problem is doing all of them, every time, in the seconds after a signal fires, when the incentive to skip checks is strongest. This is why Belphor runs them mechanically: every signal passes a rug registry of known bad mints, tickers, and deployer wallets, and live trades additionally verify authorities, pool depth, and an executable sell route before any money moves. The checks that catch rugs are boring, and boring is exactly what software is for.
If you have already been rugged
Treat it as data. Record the mint, the deployer, and the mechanism, because rug operators repeat themselves, and the wallet that rugged you will usually deploy again. Registries of rugged deployers exist precisely because recidivism is the norm in this corner of the market. And accept the base rate: most memecoins go to zero with or without a rug. Position sizing, not perfect detection, is what makes the difference survivable.
How often do Solana memecoin momentum signals survive? Death rate and hit rate by pool age, by volume acceleration and by score, measured on thousands of detected signals. Recomputed daily.
Belphor screens every signal against a live rug registry and verifies authorities, pool depth and an executable sell route before its bot trades. Every outcome is published, wins and losses alike.
Explore Belphor →