What a reshape actually is.
A reshape is the act of closing your current LP range and opening a new one. The agent withdraws your liquidity from the old bins, swaps the resulting tokens back into the right ratio for the new range, and deposits into the new bins.
From the outside it looks like one event. Under the hood it's three or four transactions, each with their own gas, their own slippage, their own chance of partial failure.
Why ranges go stale.
When the price moves, your range stops being useful. If the price walks above your range, you end up holding all of one token. You're no longer collecting fees. The pool is still trading, traders are still paying fees, but those fees are flowing to LPs whose ranges are still active. Not yours.
The bigger the move, the more stale your range gets. A SOL move of a few percent can drift you to the edge of a tight range. A move of ten percent will likely push you out completely.
Sitting in a stale range is the most common way LPs lose to the market. It's not a flashy mistake. It's just the slow drift of money sitting where it isn't working.
The cost of reshaping.
Reshaping is not free. Every reshape pays priority fees on Solana, plus the small swap slippage when rebalancing the inventory between the two tokens. On a volatile pair with a tight bin step, that swap can eat a meaningful portion of the fees you just collected.
There's also a real cost in opportunity. If you reshape mid-move, you might lock in a bad inventory ratio. If you reshape too early, before the move has settled, you'll just have to reshape again ten minutes later.
Too often is bad. Too rare is bad.
Reshaping every block burns money on gas and slippage. The fees you earn never catch up to what you're paying to chase the price. This is the failure mode where an LP looks busy and feels active but the line goes down.
Reshaping too rarely is the other failure mode. You sit in a stale range, miss the trades, and watch other LPs collect fees that should have been yours.
The right cadence depends on the pair, the volatility regime, the bin step, and the size of the position. Volatile small caps want frequent reshapes with tight ranges. Stable pairs want lazy reshapes with wider ranges. The map is different for every pool.
How Solaris balances it.
Solaris is the live anchor agent on the SOL/USDC pool. It uses a volatility-aware threshold: it only reshapes when the price has drifted far enough relative to recent volatility that the new range will earn meaningfully more than the cost of moving.
When the market is calm, Solaris sits still. When the market wakes up, Solaris moves with it. The threshold is the lever. Tighter thresholds mean more reshapes and more fee capture, with more cost. Looser thresholds mean fewer reshapes and lower cost, with less fee capture. We tune the lever per pool, not per agent type.
Why we take 20 percent of profits.
Lyra charges a 20 percent performance fee on profits. There is no management fee, no entry fee, no exit fee. If your vault doesn't make money, we don't charge anything.
That fee funds the work of running these agents. Reshape gas, RPC infrastructure, monitoring, the engineers who tune thresholds when a pool's behavior changes, the auditors we pay to look at the custody program. It also funds new agents. Every time we ship a new pool or a new strategy, the funding came from past performance.
We're transparent about this because the alternative is to hide costs in opaque APY numbers. We'd rather you see the exact split: you earn the gross yield, we take a fifth of the profit, you keep the rest. If we ever stop earning the fee, we stop charging it.