Oracles and protocol-owned liquidity

Part of a series covering all 72 Ekubo DAO proposals. This post is about liquidity the DAO puts up itself, rather than rents.

Period covered: September 2024 – May 2026.

Oracles as a public good the DAO pays for

September 2024 — DAO owned liquidity for EKUBO oracle pools, $190k allocated across seven pairs: $50k each to EKUBO/ETH, EKUBO/USDC and EKUBO/STRK, $10k each to EKUBO against USDT, wstETH, WBTC and DAI. Funded with 54,000 EKUBO, 45,000 USDC and 12 ETH.

The rationale is not trading revenue. Ekubo’s oracle extension derives prices from pool state, so an oracle is only as trustworthy as the liquidity behind it. Thin pools make cheap oracles and cheap oracles are manipulable. Paying $190k to make EKUBO price data reliable and free is infrastructure spending — the beneficiaries are integrators, lending markets and Ekubo’s own backend.

Upgrade Oracle to v1.1.0 the same month made prices safer to query on-chain, explicitly so Ekubo, Inc. could wire the oracle into the routing API (which needs prices for gas estimation) and the frontend.

The pattern repeats on Ethereum: Fund Oracle Pools on L1 in January 2025, and in February 2025 Migrate Ekubo Protocol V1 Oracle Liquidity consolidating scattered V1 positions — dust in wstETH, MKR, DAI sold down — into two deep V2 pairs, ETH/USDC and ETH/EKUBO, split 1:1.

Consolidation is the right instinct. Seven shallow oracles are worth less than two deep ones, because an oracle that can be moved is not an oracle.

Going long Ethereum with the treasury

April 2025 — Deposit DAO liquidity into the L1 deployment: 750k USDC and roughly $500k of ETH moved to mainnet for ETH-wstETH and USDC-USDT pools. Positions created at Ekubo, Inc.’s deployer address, then transferred to the DAO’s L1 owner proxy.

This is the DAO becoming its own largest LP on a chain where it had no users yet. New venues have a cold-start problem: traders don’t come without depth, and LPs don’t provide depth without volume. Protocol-owned liquidity is the DAO buying its way past that, and unlike incentives the capital is still yours afterwards.

That position then becomes the DAO’s working capital for the next year — the $100k buyback, the USDC incentive campaigns and the Recovery Fund are all funded by partially withdrawing from it and returning the remainder.

The migration into V3 took two attempts: a November 24, 2025 proposal did not pass, and a near-identical one passed a week later, moving the old positions into V3 and adding $750k to a USDC pair.

The EKUBO liquidity problem, stated plainly

April 2026 — EKUBO Liquidity Support opens with a measurement rather than a pitch: a $10k trade moved the EKUBO price by more than 15%.

It then lists what that breaks, and the list is the real argument. Holders can’t exit. Buyers can’t enter. Lending markets can’t safely integrate the token. Exchanges are less interested. And — the one that closes the loop — incentives paid in EKUBO are less attractive and don’t scale, because a reward token you can’t sell without moving the price 15% is not worth its quoted value.

That last point connects directly to the incentive campaigns, where the DAO switched from EKUBO-denominated to USDC-denominated rewards in April 2026 for exactly this reason.

DCA pools need liquidity too

Two proposals fund the TWAMM side. January 2026 allocated $120k to Ethereum DCA pools — $70k to ETH/USDC and $50k to EKUBO/USDC, both at 0.3% — so DCA orders have something to execute against. May 2026 migrated that liquidity to v3.1.1 after a flaw was found in TWAMM’s order-cancellation fee that let users circumvent it entirely; the new version removes the cancellation fee altogether, and the change was verified by an independent auditor.

Note the sequencing: the DAO funded DCA liquidity in January, discovered a fee-avoidance flaw by May, and moved the money rather than leaving it in a contract with known-wrong economics.

The proposals

Next: getting to Ethereum.