Tokenized Block

Build a block — name, symbol, decimals, supply, description, colour — and it becomes a B-20 native token on Base, with a permanent Uniswap v4 market.

Live: https://tokenizedblock.space

The block is created with no administrator and no minter: no account holds DEFAULT_ADMIN_ROLE or MINT_ROLE, at any point, including at creation. Its supply is minted in the creation transaction itself and equals the supply cap, so no further unit can ever exist. Nobody — the creator included — can mint more, or change its name, logo or description.

What this page does, and what it does not

• It presents transactions. Your wallet shows them and signs them.
• It holds no key, sends nothing on your behalf, and has no backend that can move funds.
• Every state it shows is re-read from the chain — never inferred from a transaction having been broadcast. A green result means a receipt with status 0x1 was read back.

What it costs

moment	cost	where it goes
Create	free — createB20 on the B-20 factory, value = 0	nowhere; only network gas
Launch (open the market)	0.0003 ETH life fee, paid inside inscrire	the fee wallet
each swap	0.5 % of the swap, taken by the hook	the fee wallet, in full

Creation has no service fee. An earlier version of this file claimed a 0.001 ETH create fee; the code says otherwise — utiliseCreateRouter() returns false, so Create goes straight to the factory at value zero. That claim was wrong, and it is retracted here rather than quietly deleted.

Fee wallet, and the only one: 0xa6cF99D35949c6cB911adB910078F4Ca46F0f5d4. Read it yourself — feeWallet() on the hook returns it.

The hook

Current hook: 0x5926abdAbf5D0006Ee960A8270f3e124e5a764cc (Base mainnet).

read on chain	value
HOOK_FEE()	5000 / 1e6 = 0.5 %
feeWallet()	0xa6cF…f5d4
fee split	100 % to the fee wallet — the creator's share was removed

A hook is part of the Uniswap v4 PoolKey. A market that is already open therefore keeps the hook it was opened with, forever — no migration exists, and none is claimed here.

What a block can be priced in

The hook's deviseAdmise mapping is written only in the constructor, and the contract has no setter — so the list is fixed at deployment, for that hook, permanently. Measured on the live hook on 2026-09-21:

17 currencies — native ETH, TBLOCK, USDC, cbBTC, and 13 tokenized Coinbase stocks (AAPLc, AMZNc, COINc, CRCLc, GOOGLc, INTCc, METAc, MSFTc, MSTRc, NVDAc, SNDKc, SPCXc, TSLAc).

Run it yourself: node devises-admises.mjs. It reads the mapping address by address, and also asks a witness address that has no reason to be admitted — if the witness came back admitted, the script says the table is unreliable instead of printing it.

⛔ Being priceable in a currency is not the same as being traded in it. This repository measures the first. The second is below, and it is much thinner.

Measured on Base mainnet

14 days ending 2026-09-21, 1818 log windows read, zero windows missed — so these are counts, not floors:

what	measured
fee collections	41 (2.93 / day)
ETH reaching the fee wallet	0.002798464 ETH (the wallet's share of the fee)
paid by addresses that are not ours	32.8 %, from 5 distinct addresses
which hooks collected	V1: 39 · V2: 2

And the part that is not flattering, published next to the rest:

• Two thirds of that ETH came from our own wallets trading our own blocks. Only 32.8 % is outside money.
• Hooks V6, V7 and V8 were deployed, verified on chain, and had zero markets on them. The cause was not demand: the app's fast Create path had the hook written as a constant (HOOK_V5), while only the step-by-step path read the chain. Fixed on 2026-09-21; test-un-seul-choix-de-hook.mjs now refuses a second copy of that decision.
• Total revenue to date is well under one cent per day. Nothing here is a business yet.

Verified on Base Sepolia (kept — it is what proves the encoders)

what	evidence
the creation calldata is canonical	byte-for-byte identical to what forge produces via B20FactoryLib
a B-20 can hold liquidity in a Uniswap v4 pool	positions 27264 / 27265
that pool quotes, both ways	measured curve, 0.3001 % → 83.34 %
that pool is tradeable	tx 0x2f952a47…d00ff — the quote was exact to the unit

Fork it

git clone https://github.com/Tokenized-Block/tokenizedblock-app.git
cd tokenizedblock-app
python -m http.server 8000

Static files, no build step, no dependencies. Then open http://localhost:8000/app.html.

⚠️ A wallet's built-in browser on Android refuses plain HTTP (ERR_CLEARTEXT_NOT_PERMITTED). Use the hosted HTTPS page for mobile wallets, or a desktop browser with a wallet extension.

If you fork this and open markets, the fee wallet above is what the deployed hooks pay. It is immutable inside those contracts — changing the constant in this repository changes nothing on chain. To send fees somewhere else you would have to deploy your own hook, mine its address for the permission bits, and pair your blocks with it.

Running the instruments

Every claim above has a script behind it, and each script states the bound of what it can prove.

⛔ They are not on main. This branch still powers GitHub Pages — checked against the API on 2026-09-21, it reports status: built and serves https://tokenized-block.github.io/TokenizedBlock/ — so it holds a redirect stub, not the app.

The application and every measurement script live in a separate public repository, whose main is the code running at https://tokenizedblock.space:

https://github.com/Tokenized-Block/tokenizedblock-app

Its DEPLOY.md states exactly where a fork's fees go, and why a fork cannot redirect ours.

(An earlier draft of this paragraph said Pages had been retired. It has not — the API says it is still built and serving. Corrected here rather than quietly dropped.)

git clone https://github.com/Tokenized-Block/tokenizedblock-app.git
cd tokenizedblock-app
node devises-admises.mjs
node ou-vit-le-volume.mjs 1
node reference-frais.mjs

• devises-admises.mjs — what a deployed hook accepts as a pair currency.
• ou-vit-le-volume.mjs — where B-20 swap volume actually goes: hooked pools, or free ones.
• reference-frais.mjs — fees now, against a reference frozen before the V6 hook existed, so a change can be attributed instead of assumed. It returns PAS_DE_CAUSE and publishes nothing when the hook being compared has produced no collection at all.

The test suite is 34 files; all were green on 2026-09-21.

for f in test-*.mjs; do node "$f"; done

Files

app.html the app · encodeur.js B-20 creation calldata · lancer-pool.js the launch plan · pool.js Uniswap v4 liquidity and swap · tokenomics.js hooks, supply, and the single function that chooses a hook · lecteur.js reads a block's on-chain metadata · keccak.js Keccak-256.

Every encoder is compared byte-for-byte against forge output in the development repository; the references are read from simulation artifacts, never transcribed by hand.

Method

No undated figures. No ARR, MRR or projection appears anywhere in this repository. A number that was not measured is not written down, and a measurement that went against us is printed next to one that did not.

Where B-20 volume actually lives

A partner model (Grok / "Zero 1") claimed that B-20 volume goes to unhooked pools, reached through DexScreener and aggregators rather than through any app. If that were true, the fee rate would not be the lever — discoverability would be. It is a testable claim, so it was tested.

ou-vit-le-volume.mjs reads every Initialize on the Uniswap v4 PoolManager over a window, keeps the pools with a B-20 side, then counts every Swap and splits them by hook.

Measured over 1 day ending 2026-09-21, 100 % of blocks read (zero unreadable):

	pools	swaps	distinct addresses	swaps per address
pools with no hook	20	282	19	4.3
pools with a hook	242	25 395	35	725.6

The claim is refuted, and then the answer turns out to be worse than either side expected.

Swap count is not on free pools — it is overwhelmingly on hooked ones belonging to other teams, so a hook does not repel traders. But 725 swaps per address is automation, not demand. The column that describes buyers is the address column, and it says that across the whole of Base, in a day, 54 distinct addresses touched a newly-born B-20 pool at all.

So the problem is not our fee rate, not the pair currency, and not a missing card on-ramp. There is no large pool of B-20 buyers being captured by someone else. There is barely a pool of buyers.

⛔ An earlier version of this section reported only the swap counts and concluded "B-20 tokens do get traded, at scale". That was wrong in the way that matters, and it is corrected here rather than edited away: 43 263 swaps looked like a crowd, and it was a handful of bots.

⛔ Two bounds, both load-bearing:

• Only pools born inside the window are seen. An older, active pool does not appear.
• A first version of this script read 19 windows out of 303 and would have published a ratio built on 6 % of the period — and the windows that failed were the busiest ones, so the bias ran in the direction of the conclusion. The script now splits a window in half on failure until it passes, and refuses to give a verdict at all if a single block stayed unread. That refusal is what makes the table above worth reading.