How to Spot a Rug Pull on Solana: 7 On-Chain Checks Before You Buy
You can't prove a meme token is a rug pull, only gauge the risk. 7 on-chain checks to run before buying: where to look, what each shows and what it can't.
You can’t prove a meme token is a rug pull. You can only judge how risky it is. Before buying, there are 7 things you can check on-chain: the creator’s buys and holdings, the creator’s launch history, concentration among top holders, whether there were bundled buys at launch, where top wallets got their funds, whether the social links are real, and where the liquidity sits and who still holds token authorities. These help you spot tokens with especially high risk. The reverse doesn’t hold: a token that passes every check can still go to zero.
On pump.fun, a “rug pull” usually isn’t pulled liquidity
A “rug pull” originally meant the team draining the funds from a trading pool, so buyers suddenly couldn’t sell their tokens. On pump.fun, that route is largely closed.
According to pump.fun’s public developer documentation, a new token first trades on a bonding curve, a program that quotes prices automatically by formula, not a pool funded by the creator; the only withdraw instruction in the documentation has been disabled. Once all tokens on the curve are bought out, anyone can trigger the migration that moves liquidity into a PumpSwap pool, and the LP tokens received from the migration are then burned. Think of LP tokens as the pool’s “deposit receipt”: whoever holds them can withdraw the money in the pool. Once burned, nobody can.
So for a token launched on pump.fun that graduates normally, a so-called rug usually looks different: after the price has been pushed up, the creator and wallets linked to them sell their cheap tokens in bulk. The pool is still there; the SOL in it has been swapped out by them. Most of the 7 signals below are about finding out “who holds cheap tokens, and are they working together”.
The 7 signals at a glance
| Signal | Where to check | What it can show | What it can’t show |
|---|---|---|---|
| 1. Creator’s first buy and holdings | Solscan token page, Transaction and Holders tabs | How many cheap tokens the creator holds | Whether they moved tokens to other wallets |
| 2. Creator’s launch history | Creator’s account page | Whether the address launches in bulk | The real background of a new address |
| 3. Top-holder concentration | Holders tab, Analysis tab | Whether a few addresses could crash the price at once | Whether scattered addresses are the same person |
| 4. Bundled buys at launch | Earliest trades on the token page | Whether a batch of wallets grabbed launch supply at the same moment | Whether those wallets are the same person |
| 5. Funding source of top wallets | SOL transfers on each wallet’s account page | Whether several wallets were funded from the same source | When the source is an exchange, the trail ends |
| 6. Whether social links are real | The links on the token page themselves | Fake or borrowed links mean clearly higher risk | Real links don’t make it trustworthy |
| 7. Liquidity and token authorities | Token page overview and market info | Whether there is a pool that can be drained, or authority to mint more | The risk of holders dumping |
How to check each signal
Every check starts at solscan.io: paste the contract address (mint address, also called the CA) into the search box and open the token page.
1. Creator’s first buy and holdings
pump.fun’s developer documentation states that the creator can buy in the same transaction that creates the token. On the token page’s Transaction tab, scroll to the earliest record and see whether the creation transaction included a buy and how large it was. Then check the Holders tab for how much the creator address still holds.
A large buy at creation that hasn’t been sold means the creator is sitting on a big block of the cheapest tokens. But if they have moved tokens to other wallets, Holders won’t show it. Go back to their account page and check outgoing transfers.
2. Creator’s launch history
Open the creator address and count the token-creation operations in its transaction history. An address that launched hundreds or thousands of tokens in a short time can hardly be expected to take responsibility for any one of them. For the step-by-step method, see Why one wallet launches hundreds of tokens on pump.fun. A new address that has launched only one token may also be an old hand using a fresh alias.
3. Top-holder concentration
The Holders tab lists each holder’s address and its share of total supply. The Analysis tab has a pie chart where you can view the combined share of the top 10, 20, 50 or 100.
Remove non-individual accounts before you calculate. Before graduation, unsold tokens sit in the bonding-curve account; after graduation, the PumpSwap pool account holds a large block. Neither is a “person” holding tokens. After removing them, if the top few individual addresses together hold a high share, any one of them selling would hit the price hard.
The reverse is not proof of safety either. One person can split tokens across dozens of wallets, and on this table that looks like dozens of “different people”. That is why you go on to checks 4 and 5.
4. Bundled buys at launch
Bundled buys mean one person using a batch of wallets to buy at the same moment at launch, sometimes in the same block, so the supply looks spread across many addresses.
To check, look at the earliest few dozen trades on the token page: every entry on the Transaction tab shows the block number and time. If a string of different wallets bought in the block containing the creation transaction, or in the few seconds right after, and those wallets are still among the top holders, be on guard.
Unrelated automated buying programs can also squeeze into the same block at launch, so buying in the same block does not by itself mean one person. Read it together with the next check.
5. Funding source of top wallets
Open the account pages of the top few wallets and look at the senders in their SOL transfers. If several wallets all received SOL from the same address shortly before buying, they are likely controlled by the same party, and the “spread” you saw earlier counts for much less.
If the sender is an exchange hot wallet, the trail ends there, because exchanges pay out to huge numbers of unrelated users. Solscan tags some known addresses with exchange names. When you see such a tag, you know this is where to stop.
6. Whether social links are real
The website, X and Telegram links on a token page are entered by whoever launched the token. pump.fun’s terms of service state that the platform is not responsible for user-provided content, and require users to do their own due diligence on tokens and creators.
Open each one. Is the X account brand new or recently renamed, and do its posts include this token’s contract address? Is the website a single page? Do the links borrow a well-known account or another project? Fake or borrowed links mean clearly higher risk. Real links only show that someone made a little effort. A decent-looking website and account don’t take long to set up.
7. Liquidity and token authorities
The token page’s market info lists which platforms and trading pairs hold liquidity for the token. For a token that graduated normally, the LP tokens of the main PumpSwap pool have been burned. But the official fee page also mentions that PumpSwap can have “non-canonical” pools, and other platforms can create pools too. The money in those pools belongs to whoever deposited it, and they can withdraw it at any time.
The token page overview also shows the mint authority. According to Solana’s official documentation, the account holding it can mint new tokens. The other authority, the freeze authority, can stop tokens in a given account from being transferred. If either is still held by someone, they have a way to dilute or lock your tokens. If neither exists, that only rules out these two tactics.
Why you can still lose everything after checking all 7
Every check above has a workaround: launch from a new address, spread tokens across wallets, fund through an exchange, pay for accounts. The chain records what happened, not who is working with whom. However carefully you look, you only see the part they failed to hide.
Even if nobody behind a token is acting in bad faith, its price depends only on later buyers being willing to pay more. Once buying stops, it falls. Fees count too: according to the official fee page, each trade in the bonding-curve stage pays 1.25%, and orders placed through the pump.fun website or app pay an additional platform fee.
These 7 signals can help you filter out tokens with especially obvious risk, but they can’t find you a “safe” one. No single signal, and no combination of them, can guarantee that a meme token won’t rug.
All of the above has to be checked one token at a time. To get a sense of the overall environment first, such as whether there are unusually many new tokens today or serial deployers are unusually active, take a look at the daily report.
References
- pump.fun developer docs, Pump Program: https://github.com/pump-fun/pump-public-docs/blob/main/docs/PUMP_PROGRAM_README.md
- pump.fun developer docs, coin creation: https://github.com/pump-fun/pump-public-docs/blob/main/docs/instructions/COIN_CREATION.md
- pump.fun fees: https://pump.fun/docs/fees
- pump.fun terms and conditions: https://pump.fun/docs/terms-and-conditions
- Solscan docs, Token page: https://docs.solscan.io/browsing-the-site/token-page
- Solscan docs, Labeling: https://docs.solscan.io/transaction-details/labeling
- Solana documentation, Tokens: https://solana.com/docs/tokens
