Introduction

About Optionbolt

Optionbolt is permissionless, capital-efficient options on Hyperliquid: a writer backs an option with a perp position they already hold and a small refundable bond, and a neutral clearing core collateralizes and settles it while the market prices the premium anywhere. This page is the why behind it, and the path that got here.

The last missing primitive

DeFi rebuilt finance one primitive at a time, and each one took something that used to need permission and handed it to anyone. Uniswap did it for market making, Aave for lending, perpetual futures for leverage, stablecoins for dollars, and real-world assets are doing it for equities. Exchange, lending, leverage, currency, equities: a whole financial system rebuilt so no one has to ask. One primitive never crossed over. Options.

Serious teams tried, and they hit a wall with two sides. Fully collateralize every option and it is safe but starves: locking a whole coin to write one call is capital almost no one leaves sitting idle, so it cannot compete on price. Post less than you owe and you have built a machine: partial collateral means the protocol has to know when a writer is underwater, which needs a price, which needs an oracle, which needs a liquidation engine, which needs an insurance fund for when that does not clear. The teams that built that machine built it well. What killed them was welding it into the core, one frozen vertical that could never take on the next idea (a new underlying, a better venue, an app plugging in as a broker). Off-chain, options run as a stack: a neutral clearer at the bottom, exchanges above it, brokers on top. No one on-chain ever built the bottom layer, so nothing above it could grow.

Where I started

I started at the other end of that wall. I built a fully-collateralized, oracle-less options exchange on Polygon: buy calls and puts, write them covered or cash-secured, settle by delivering the real asset, with an order book in the shape of Hyperliquid’s or Polymarket’s. It reads no price and never liquidates, because there is nothing to liquidate when the collateral is locked up front. And it is genuinely good at the one thing nobody else can touch: assets that do not move, they jump. A prediction share sits at a penny for weeks, then a race is called and it is a dollar before the next block, with nothing in between. No oracle can price that and no liquidation can act inside a single-block gap, so every partial-collateral protocol is insolvent the tick it happens. Mine does not flinch, the collateral was locked before the jump and it is still locked after.

But that strength is the ceiling too. Oracle-less, fully-collateralized options are perfect for the long tail, the jumpy, hard-to-price assets, and that is a real but small market. The money, and the fight worth having, is capital efficiency on the liquid assets: ETH, BTC, and increasingly the oil, the equities, the forex, the indices that Hyperliquid is putting on-chain to compete with traditional finance directly. Competing with tradfi on the assets it used to have a moat on is the end state I want for on-chain options, and oracle-less cannot get there. So I moved.

On-chain options spent years trying to beat perps, one more way to place a leveraged bet, and lost, because perps are simpler, deeper, and the biggest thing crypto has built. But perps were never the competition. They are the collateral. A perp trader already holds leveraged exposure, tens of billions of it, and until now not a dollar of it could be written against an option. Optionbolt turns that exposure into the backing for a capital-efficient option: post margin, not the full notional, and the perp you are already comfortable holding becomes the cover. Hyperliquid is where that exposure lives and where the tradfi-competing markets are being built, so it is where this belongs.

The neutrality I learned the hard way

The other lesson I learned the expensive way. I built a parlay engine on top of Polymarket, and Polymarket built its own. I built options clearing that leaned on Polymarket’s low fees, and Polymarket raised them past 1% on the markets that mattered. It is the lesson the last generation of DeFi keeps teaching: platforms wear neutrality as a costume long enough to take the users, then cash it in. Build on a platform that can copy you or tax you, and you are building on someone else’s terms.

So Optionbolt is built to be the neutral layer, not a platform anyone has to trust. The core reads no oracle of its own and runs no liquidation engine of its own; it leans on Hyperliquid’s, so nothing about your risk sits on a private server I control. The fee split is hardcoded and uniform, and a fork clears through the same core and pays the same split, so fairness is structural, not a promise I am asking you to believe. Governance is deliberately minimal, only the conservative volatility floors that keep the system solvent, the same parameter-risk a lending protocol like Aave governs, and it is headed to a DAO. The whole point is to build the floor and not compete with the people who build on it.

What Optionbolt is

Concretely, Optionbolt is a clearing core for perp-backed options on Hyperliquid. A writer backs an option with their own Hyperliquid perp and posts a small refundable bond; the core enforces that the cover and the bond always stand behind the option, and cash-settles it when it is exercised, expires, or the cover is force-closed. HyperCore does the liquidation natively. The premium is priced and sold wherever, the core never runs price discovery or holds it. Collateral is floored by a conservative, governance-set volatility model anyone can verify, so there is no live vol feed to depend on and nothing to fail closed, and any asset Hyperliquid lists is listable here.

Because the core only clears, it is a primitive others build on rather than a closed exchange. Any venue can plug in: Optionbolt’s own exchange, a third-party order book, an AMM, an OTC desk, or a protocol that has never listed an option before. A lending market could let depositors write covered options against collateral they already hold, or take option positions as collateral. The core is neutral infrastructure, and any exchange, Optionbolt’s own included, is a consumer of it, not a gate in front of it.

The stack

Optionbolt builds the full stack to bootstrap the market, on terms hardcoded to stay fair. The core is the neutral primitive; everything above it is ours to start and open for anyone to replace.

• Clearing core. Settles every trade. Perp-backed, cash-settled, with no oracle or liquidation engine of its own. See the Clearing overview.
• Exchange. Where options trade on Optionbolt: open-source, fork-invited, and one of many venues that clear through the core. Builders earn builder-fee attribution for the order flow they route. See the CLOB overview.
• Frontend. One reference interface: trade on it, fork it, or build your own.
• Governance. The entities and process behind the protocol. See Governance.

The core-and-matcher fee split is hardcoded in the contract and uniform. A fork still clears through the core and pays the same split, so fairness is structural, not a promise.

What you can trade

• Calls and puts, backed by a Hyperliquid perp for a fraction of the notional. A 1x perp fully backs an option; leverage is simply how a writer writes cheaper, not a requirement.
• Cash-secured puts, for a writer who would rather lock the stablecoin than run a perp (and the way to sell a call and a put on the same asset at once).
• Credit and debit spreads, cash-settled and defined-risk. A credit spread locks the width between the strikes as collateral; a debit spread only costs the premium you pay.
• Asset-agnostic, starting with ETH, then BTC, and eventually every liquid Hyperliquid market.

Where to go next