What the agent watches
Around the clock, the agent monitors every position’s place in its range, the pool’s realised volatility, current gas and swap costs, and the value of the emissions at stake. From these it decides whether and when re-centring is worth it.The core judgement
Rebalancing is not free — it costs gas and swap impact, and doing it badly is the classic way LPs bleed. The agent’s rebalancing policy weighs the cost of acting against the cost of sitting idle, per position, using decision rules tuned on live production data. The exact triggers are closed source — they are among the most valuable things we operate, and publishing them would also make the agent’s behaviour front-runnable. What we commit to publicly:- No rebalancing on a timer. Every action is justified by position state and economics, never by a schedule.
- Every rebalance is visible. Each one appears in your statement with the transaction, the drift that prompted it, and the new range.
- Costs are booked against the strategy. Gas and swap costs of tending are counted in your strategy delta — the agent’s performance is measured net of its own activity, so it has nothing to hide behind.