Your portfolio shouldn't be a public wallet.
Shielded pool for SOL on Solana live on mainnet
https://t.co/RtTz0F6080
4c1XZRqFV6y8pAckHru5oiGPYUFw1eQ3kFotPLHrpump
We’ve been pushing our independently built, clean-room @solana validator client pretty hard.
PoH: 4.8 ms
Verification: 6.4 ms
SHA-256: 9.2 ns/hash
Replay: ~78 blocks/sec
Identical final bank hash.
Finalizing the client soon. Still more performance on the table.
Three things shipped since yesterday, all in the repo.
1. The token does something now. A deposit pays 0.5%, held on its own address.
Burn KEEPHOST and you get a key: that wallet never pays the fee again. Those fees pay the relayer at withdrawal — so a withdrawal returns the whole denomination, funded by the people who did not burn.
2. Any SPL token can have a pool. Same tree, same circuit, same proof, same relayer; the vault becomes a token account owned by an address with no private key. Tested with a mint shaped like USDC.
3. The program went from 518 kB to 437 kB by talking to the token program directly instead of through a library. On Solana that is rent paid for ever it halved the cost of the upgrade.
Every line of it passes an end-to-end test against the exact binary that will be deployed: deposit, proof, withdrawal, and the same receipt refused twice.
https://t.co/KnDFcmqvhE
USDC pools: written, tested, not open.
Same proof as the SOL pools the vault just becomes a token account owned by an address with no private key. Three deposits in, one withdrawal out to a fresh account, the same receipt refused twice.
It opens with the next deploy.
That deploy waits on the trusted setup, and the setup waits on contributors, not on me.
Three things shipped since yesterday, all in the repo.
1. The token does something now. A deposit pays 0.5%, held on its own address.
Burn KEEPHOST and you get a key: that wallet never pays the fee again. Those fees pay the relayer at withdrawal — so a withdrawal returns the whole denomination, funded by the people who did not burn.
2. Any SPL token can have a pool. Same tree, same circuit, same proof, same relayer; the vault becomes a token account owned by an address with no private key. Tested with a mint shaped like USDC.
3. The program went from 518 kB to 437 kB by talking to the token program directly instead of through a library. On Solana that is rent paid for ever it halved the cost of the upgrade.
Every line of it passes an end-to-end test against the exact binary that will be deployed: deposit, proof, withdrawal, and the same receipt refused twice.
https://t.co/KnDFcmqvhE
36 hours in: mainnet program, browser deposits and withdrawals, a relayer, the source public with a green CI, and SPL pools tested and waiting on the next
deploy.
Most projects are still writing a whitepaper at this point.
The slow part left is the trusted setup, and that one is slow on purpose.
Your portfolio shouldn't be a public wallet.
Shielded pool for SOL on Solana live on mainnet
https://t.co/RtTz0F6080
4c1XZRqFV6y8pAckHru5oiGPYUFw1eQ3kFotPLHrpump
36 hours in: mainnet program, browser deposits and withdrawals, a relayer, the source public with a green CI, and SPL pools tested and waiting on the next
deploy.
Most projects are still writing a whitepaper at this point.
The slow part left is the trusted setup, and that one is slow on purpose.
36 hours in: mainnet program, browser deposits and withdrawals, a relayer, the source public with a green CI, and SPL pools tested and waiting on the next
deploy.
Most projects are still writing a whitepaper at this point.
The slow part left is the trusted setup, and that one is slow on purpose.
USDC pools: written, tested, not open.
Same proof as the SOL pools the vault just becomes a token account owned by an address with no private key. Three deposits in, one withdrawal out to a fresh account, the same receipt refused twice.
It opens with the next deploy.
That deploy waits on the trusted setup, and the setup waits on contributors, not on me.
USDC is next, and the code for it is already written and tested.
A KeepHost pool can now hold any SPL token. Same tree, same circuit, same proof, same relayer the vault becomes a token account owned by an address with no private key, and that is the entire difference.
Tested against a mint shaped like USDC: three deposits in, a withdrawal out to a fresh account, the same receipt refused the second time.
Not open yet. It opens with the next deploy, which waits on the trusted setup being redone in public and that one needs contributors, not buyers.
https://t.co/KnDFcmqvhE
KeepHost now takes any SPL token, not just SOL.
Same tree, same circuit, same proof, same relayer.
The only difference is that the vault is a token account owned by an address with no private key, instead
of an address with no private key holding lamports.
Tested against a mint shaped like USDC: three deposits in, a withdrawal out to a fresh token account, and the same receipt refused the second time.
This is what the front page owed. It promised tokenised assets while the pool
took SOL and nothing else that gap is closed in code, and the pools open with the next deploy.
https://t.co/KnDFcmqvhE
KeepHost now takes any SPL token, not just SOL.
Same tree, same circuit, same proof, same relayer.
The only difference is that the vault is a token account owned by an address with no private key, instead
of an address with no private key holding lamports.
Tested against a mint shaped like USDC: three deposits in, a withdrawal out to a fresh token account, and the same receipt refused the second time.
This is what the front page owed. It promised tokenised assets while the pool
took SOL and nothing else that gap is closed in code, and the pools open with the next deploy.
https://t.co/KnDFcmqvhE
Your portfolio shouldn't be a public wallet.
Shielded pool for SOL on Solana live on mainnet
CA : 4c1XZRqFV6y8pAckHru5oiGPYUFw1eQ3kFotPLHrpump
Visit : https://t.co/RtTz0F6xXy
👨💻 Newsletter #2 is coming with next layer
Seven deposits went into the 0.1 SOL pool. Two came back out, on addresses with no history.
Go look. The vault is public, the payouts are public, the amounts are identical by design.
Match one to the other you can't, and neither can I.
That is the whole product, running on mainnet, in a pool small enough that you can count it by hand.
Visit : https://t.co/r6TuofZqxT
👨💻 Newsletter #2 is coming with next layer
Seven deposits went into the 0.1 SOL pool. Two came back out, on addresses with no history.
Go look. The vault is public, the payouts are public, the amounts are identical by design.
Match one to the other you can't, and neither can I.
That is the whole product, running on mainnet, in a pool small enough that you can count it by hand.
Visit : https://t.co/r6TuofZqxT
Your $ZEC can stay shielded, and you can still see exactly what you're approving 🔐
Shielded @Zcash is now supported in Ledger Wallet, with every transaction approved on your Ledger signer.
Shielded or transparent: how are you holding your Zcash? 👇
The book on Solana is public. KeepHost is the layer that stops it.
Fixed sizes. A vault with no key. A Groth16 proof that spends one deposit without naming it. A relayer so the destination is not the wallet that left. That path already ran on mainnet.
CTHg29kf7L6TNDH5TSd3tdoZfsmP39JjyQWKmPtEY1YW
What is coming same circuit, harder guarantees.
SPL in the pool. USDC first, then the tickers that already trade here.
Association sets: you prove you are in a set you chose, not in “everyone who ever deposited.” First-gen shielded pools died because they were only a crowd of everyone. This is how you build the same math without that stain.
More than one relayer, picked by the page.
If you write programs on this chain: open the site, walk the two tabs, match source to bytecode when the repo is up.
Break it in public.
https://t.co/0eFWiCDLy2
CA: 4c1XZRqFV6y8pAckHru5oiGPYUFw1eQ3kFotPLHrpump