For about 10 hours and 22 minutes on 19 July 2026, this address repeatedly opened, reduced, and closed concentrated-liquidity positions. Across its complete 694-transaction history, each of those three behaviors appeared exactly 186 times; another 58 transactions executed asserted swap-route patterns. The balance resembles systematic liquidity-position maintenance, possibly automated, but the evidence does not establish who operated the address or whether the activity was profitable.
Rhythm and repertoire
The whole-history median gap was 2.937 seconds, although long pauses raised the mean to 53.857 seconds. The longest idle period was about 50 minutes 34 seconds, and the busiest minute—21:38 UTC—contained 18 transactions. The longest consecutive run of successful executions was 216 transactions. In the capped latest-500 stream, the median gap was 2.579 seconds, covering activity from 16:16:01 to 23:25:47 UTC.
The three liquidity behaviors formed a remarkably balanced repertoire: 186 openings, 186 reductions, and 186 closures, each 26.8% of the complete history. The opening patterns prepare funding through coin splits and a fixed-amount gateway before delivering a position. Reduction patterns remove liquidity and collect fees or rewards while leaving the position open. Closure patterns remove remaining liquidity, require the position to be empty, consume it, and deliver the resulting balances.
The 58 swap-route transactions were the main deviation. Their routes used router contexts and Bluefin or Cetus legs, followed by settlement and sometimes an asset transfer. That call structure establishes route execution, not necessarily a completed exchange.
Three transactions that show the difference
Hz6vDwMa is a representative latest-500 reduction. It ran a liquidity-removal call followed by fee collection. At the transaction boundary, the sender's movements were +146.356948 USDSUI and +0.003727352 SUI; the record does not attribute those amounts to a specific call.
AB3AMhBR shows the swap-shaped exception. It routed value through eight connected calls, including four exchange-related calls, then reached router settlement. The boundary movements were −50.480566465 SUI and +37.833072 USDSUI, but the stored explanation explicitly leaves the exchange outcome unresolved.
BpnAdWcJ shows the failure mode. It opened a position and prepared two supporting values before the liquidity-addition call aborted at utils::deposit_balance. Application effects rolled back, so the reached opening call does not prove that a position was funded. The transaction recorded −0.000181472 SUI in both the holding and gas evidence; those lanes describe overlapping evidence and must not be added together.
What moved
Across the complete history, the holding evidence shows gross SUI inflows of 18,470.856979964 SUI and outflows of 18,485.599530972 SUI, for a net holding movement of −14.742551008 SUI. USDSUI shows 14,696.694362 in and the same amount out, for a net of zero. There was also a holding inflow of 0.003537 USDC and 0.041764 DEEP, each in one transaction.
The separate gas evidence shows 2.105383220 SUI out and 1.691752392 SUI in, for a net gas movement of −0.413630828 SUI. All 694 transactions used this same address as gas owner, so none were sponsored. Gas and holding figures are reported separately because the same SUI can appear in both evidence lanes. These movements establish bounded sender-side flows, not prices, profit, or loss.
Eleven of the 694 transactions failed. Six stopped at clmm_math::get_liquidity_by_amount and five at utils::deposit_balance. The complete record contains movement evidence for every transaction, with no omitted movement sources. The detailed transaction stream is capped at the latest 500 rows, so its five failures and 55 transactions without an asserted behavior should not be read as complete-history counts.
The useful exits are AB3AMhBR, Hz6vDwMa, and BpnAdWcJ; the recurring mechanisms are sui/open-liquidity-position/roster-f0024/v1, sui/decrease-liquidity/packed-e118-a5a38aa8fed1-18f1face8b33/v1, and sui/execute-swap-route/packed-e118-617f2e2506de-7cd3339cacf/v1. Together they show a highly repetitive position-management routine with a smaller, more difficult-to-resolve route-execution branch—not an identified person, company, bot, or proven trading result.