Earn
The unified earn surface — lending supply sides and vaults as one row shape — and per-row history / withdrawability metrics
The unified earn listing
Lending supply sides and vaults as ONE paginated row shape (`EarnMarket`), every rate in PERCENT (`rateUnit: 'percent'`). Served from the origin's `/earn/latest` when it can answer, else from an edge merge of `/pools/latest` + `/vaults/latest`; the shape is identical either way and a failing source is reported in `sources[]`.
Desk facet: money → credit desk → strategies
One entry per (credit desk, denomination): the tokens that put you on that desk, what DEPOSITING them pays, and what LOOPING them pays. Built from the same data as `/v1/data/earn` (its cached catalogue, with its default exclusions — `deposit.count` is what a bare listing request shows) and `/v1/data/lending/pairs/optimize` (at its defaults). Cached 5 min, served stale while refreshing.
An account's earn positions, every chain, both venue kinds
Lending accounts (one row per chain × lender, markets as `legs`) and vault share balances (one row each), with totals. A lender with isolated sub-accounts (Euler, Dolomite, Fluid / Jupiter Lend NFTs, Kamino obligations …) lists each in `subAccounts[]`: `accountId` is what the action routes take, `label` the venue's own name where it has one, `accountAddress` the address it acts as on chain, and `unsupported` — when set — why the action routes cannot build for it (a Kamino Multiply / Leverage obligation).
Withdrawability + realized rate for one earn row
**How likely is it that funds in this row get locked up — and what did it actually pay?** One market or vault, one window, measured from the recorded hourly series rather than quoted.