OOZE
MEV AND BUNDLER RETARDANT.
MULTIDIMENSIONAL. APPLY TO BLOCKCHAIN.
§ THE PROBLEM
Solana validators sort transactions by priority fee.
The richest wallet always wins.
98%
of validators run Jito's client
770/773
monopoly on block production
85%
decline in active validators since Oct 2024
$14.7M
extracted from retail in a single peak day
Bundlers pay astronomical priority fees to land in positions 1–5. They buy the float at floor prices.
By the time retail's transaction lands at position 7, the price has 10x'd. Then the dump.
This pattern runs thousands of times per day. It's not a bug — it's the mechanism.
§ WHAT WE BUILT
LIVE
Forensics
ooze.run
On-chain analysis tool. Paste any Solana token, expose bundler clustering and sandwich patterns by wallet. The proof that the problem exists.
Analyze a token →
SHIPPED
Validator Client
github.com/SwagasaurusFLEX/ooze-validator
Agave fork with a new transaction scheduler. Replaces priority-fee ordering with VRF-seeded randomness. Bundles can't be guaranteed. Sandwiches can't be constructed.
View the code →
LIVE
Validator Economics
dashboard.ooze.run
Live dashboard analyzing 773 validators across 15GB+ of on-chain data. The structural case for fair-ordering infrastructure, in numbers.
Open dashboard →
LIVE
Labs
lab.ooze.run
A playground to test Ooze. Drip free SOL, point your wallet at the RPC, send transactions through a fair-ordering validator. More tools coming.
Enter the lab →
§ CONSENSUS
Ooze doesn't break Solana. It just stops Solana from breaking users.
VALID BLOCKS
Solana's consensus rules don't constrain transaction ordering — they require valid signatures, correct state transitions, and timely propagation. Ooze produces blocks that satisfy all three. Any Solana validator (Jito, agave, Firedancer, or another Ooze node) accepts an Ooze block the same way it accepts any other.
REVENUE POTENTIAL
Bundlers pay Jito validators to extract value from users — and pools, retail, and Solana itself notice. Ooze validators capture base fees, priority fees, and inflation rewards while charging higher commission rates. They earn stake delegation from LSTs and holders who want their SOL working for the chain instead of against it. As trust in MEV-extractive infrastructure erodes, fair-ordering validators stand to take both market share and premium pricing.
§ HOW IT WORKS
01
Buffer
Pending transactions accumulate in the validator's scheduler. Wait for at least 16 transactions or 50ms.
02
VRF
Compute a verifiable random function over (slot, hash of tx_set) using the validator's identity Ed25519 keypair.
03
Shuffle
Seed a ChaCha20 CSPRNG with the VRF output. Use it to randomize transaction order before block construction.
04
Verify
Anyone can take the published VRF proof and confirm the validator didn't cheat. No new key registry. No trust assumptions.
WHAT BREAKS
Atomic bundling · Sandwich attacks · Sniper rugs · Coordinated launches
WHAT STAYS
Validator fee revenue · Honest arbitrage · Liquidations · Network throughput
§ ROADMAP
v0
SHIPPED
Validator client fork · Standalone VRF crate · Forensics tool live · Demo bundler · Comparison candlestick chart
v1
NEXT
Public devnet deployment · oozeSOL liquid staking pool · DEX integrations · Forensics counterfactual replay
v1.5
SOON
ECVRF migration · On-chain VRF proof publication · Validator monitoring dashboard
v2
FUTURE
Threshold VRF · Distributed randomness beacon · Protocol-level enforcement · Quantum entropy research