Between 19:44:00 and 19:44:59 UTC, Sui recorded 1,343 eligible transactions. One sender, 0xcde6dbe01902be1f200ff03dbbd149e586847be8cee15235f82750d9b06c0e04, accounted for 520 of them—38.7% of the minute—and paid its own gas throughout. Its activity was unusually repetitive: 290 cancellation attempts and 230 limit-order placements, all without a recorded failure. The cadence is consistent with high-throughput automated execution, but the address alone does not establish a bot or a shared operator.
The dominant pattern was DeepBook order management. Across the 1,330 transactions with a usable semantic classification, cancellations accounted for 329 rows (24.7%) and limit-order placements for 314 (23.6%). The concentrated sender used the ordinary owner-proof pattern 478 times and a repeated owner-proof placement variant 42 times. Its transactions appeared throughout the minute: the median gap was 0 ms, 327 of its 519 observed gaps were zero, and its longest gap was only 1,160 ms.
Two final transactions make the loop concrete. Aj166PFNfRys3Mxm4GhqVt1YGK1nkphpXjZzwnx25789 ran an owner-proof cancellation in the NS/SUI book. Its detail identifies an order at a 0.0154 SUI-per-NS resting price and 872.9 NS of resting size removed; that is evidence of cancellation, not evidence that the order ever traded. Moments later, B33bvkNzXxtgVPMWucZsPqrXo9cUVqTJ1FoXx1URYboH submitted a 1,050.5 NS limit order at 0.0153 SUI per NS, described as remaining on the book. These rows look like maintenance around resting orders, not a proven sequence of fills.
The minute was not only DeepBook. Price-related calls made up 219 of 1,330 semantically classified rows (16.5%), while campaign activity records made up another 148 (11.1%). In 6evUuG1SsPr856i6qFLGZ2G95JvZ9uduBTLiEWBEs61W, an oracle workflow submitted prices of 0.747278808183074 USDC per SUI and 1.338456288529224 SUI per USDC with explicit expiry. The calls ran, but the evidence does not establish that either stored price changed.
Failures were comparatively scarce: 39 of 1,343 eligible transactions failed (2.9%), below the frozen corpus rate of 8.36% across 1,000,000 transactions. Sixteen failures had no more specific recorded stop location; nine stopped at balance::split, four at a vault deposit call, and the remaining ten were spread across swap, reward, router, and slippage checks. None came from the concentrated sender, so the minute's defining feature was repetition without visible breakdown.
The movement record adds scale but not profit. Sender-boundary SUI movements in the holding lane totaled 36,010.288261384 SUI in and 35,641.201777055 SUI out, a net of 369.086484329 SUI across 484 transactions. Gas is separate: the gas lane recorded 1.306160948 SUI net out across 1,074 transactions. These lanes must not be added, and no prices or cross-asset valuation are available.
For a closer look, start with Aj166PFNfRys3Mxm4GhqVt1YGK1nkphpXjZzwnx25789, B33bvkNzXxtgVPMWucZsPqrXo9cUVqTJ1FoXx1URYboH, or 6evUuG1SsPr856i6qFLGZ2G95JvZ9uduBTLiEWBEs61W; the recurring patterns are sui/cancel-order/deepbook-owner-proof/v1 and sui/place-limit-orders/deepbook-owner-proof/v1.