Everything that can
go wrong with this.

What this is

A pump.fun coin with a program running against it. The program claims the creator fees the coin generates, spends them buying the coin back in fixed 0.1 SOL lots, and destroys everything each lot buys inside the same transaction. That is the whole of it. There is no product, no revenue, no company, no team, and nothing being built.

It is a plain pump.fun mint with no special properties, and buying it makes you an owner of nothing except the token.

What the token entitles you to

Nothing. It carries no claim on the fees, no claim on the wallet, no governance, no share of anything, and no promise that the program keeps running for any length of time. The burn destroys supply, which changes a ratio; it does not send you anything and it never will.

How it can fail

1. the fees can simply not arrive

   Creator fees are a share of volume. No volume, no fees, no claim, nothing to spend, and the tap sits shut indefinitely. Most coins do exactly this within an hour or two of launching. A shut tap is the mechanism working correctly and it is also what a dead coin looks like, and from the outside those are the same picture.

2. a buy can revert

   The price moves between quoting a lot and the transaction landing. When the move is bigger than the slippage bound the program refuses it, the lot fails, and the fee for that attempt is gone. This is ordinary and it is expected; the ledger records it as failed and the budget is not debited. Repeated identical failures raise an alarm the site renders, because that is a broken mechanism rather than bad luck.

3. the engine runs on a computer that can be turned off

   It is not a smart contract and I have never said it was. It is a program on a machine, and if that machine is off, or the network is down, or the process crashed, then nothing claims and nothing burns until it comes back. The ledger stops where it stopped and you can see that it stopped, which is more than most arrangements offer.

4. the wallet holds real money and I control it

   The wallet that claims the fees is one I hold the key to. The program will not spend more than it claimed, and the ledger and the chain both show what it did. That is a constraint on the program and not on me, and you should price that in rather than take my word for the difference.

5. the burn is smaller than it sounds

   What gets destroyed is whatever a twentieth of a SOL buys at the price at that second. Early, on a thin curve, that is a lot of tokens; later it is far fewer for the same money. A destroyed figure that keeps climbing is not the same as one that keeps mattering, and the percentage of supply next to it is the honest version of the number.

6. it buys high by construction

   The vault is fattest exactly when the coin is dearest, because fee income is volume and volume peaks with the price, which means the biggest lots are reliably the worst-priced ones. Anyone telling you their version times the market is describing something they have not measured.

What has actually been proven, and what has not

Being specific about this, because "it works" is not a claim anyone should accept without the workings.

Proven, on chain, with real money. The claim path and the buy path ran end to end on a previous coin. Twenty claims confirmed for 3.5314 SOL. Thirty-six buys confirmed for 3.3871 SOL. Six buys failed at simulation, before signing, so they cost nothing. It closed holding 0.1443 SOL that had been claimed and not yet spent.

Six failures out of forty-two attempts is what normal looks like on a moving chart, and it is not a fault.

And the run before that one failed completely. Thirteen claims confirmed for 2.3091 SOL, one hundred and one buys attempted, one hundred and one buys failed, nothing spent and nothing bought. The token program was assumed instead of being read off the mint, so every buy died at the same instruction while the claims kept confirming on top of it and the whole thing looked busy. That is the record too, and it is why the checks in front of this launch refuse to pass against a substitute coin.

Not proven. This coin. It is a new mint with a new dev wallet, so a different creator vault and a different bonding curve, and none of that wiring has ever been exercised. It gets checked against the real mint before anything is announced, and the checks refuse to pass against a substitute coin, because a dry run against a different coin proves nothing about this one.

Newest of all. Destroying the tokens inside the buy transaction. One transaction means either both instructions happened or neither did, which is the correct shape and the one that makes a revert cost a fee instead of leaving loose tokens in a wallet for somebody to take. It has not run on this coin yet either. The first lots are the real test and you will watch them at the same time I do.

The page itself

The market figures come from Dexscreener, read by your own browser every eight seconds. The ledger comes from the engine over a tunnel. Anything neither of them can supply renders as a dash, and there is no mode in which this site invents a number to fill a gap. If it is showing you a figure, something reported it.

Nothing on the front page is animated. The tap is a still image and the only things that change are figures that came off one of those two feeds.