A price-feed metronome
For the complete frozen history, 0x782439…873aa3d sent 21,019 transactions between 13:01:51 and 23:37:38 UTC. Almost all of them—21,016—ran the same verification-first price-update workflow. Every recorded transaction completed without an execution failure. The pattern is consistent with automated price-feed maintenance, but the chain evidence does not establish who operated the address or whether any resulting price was correct or economically useful.
The rhythm is striking. Across the full history, the median gap was 0 milliseconds because many transactions arrived in tightly packed batches; the mean gap was about 1.81 seconds, the 90th-percentile gap was 10.643 seconds, and the longest idle period was 21.38 seconds. The activity remained almost unchanged throughout the window: every hourly bucket was dominated by update-price, with only one transfer and two transactions carrying no asserted behavior. This is a cadence and repetition finding, not proof of a human, bot, or shared operator.
The repeated act
The reusable pattern sui/update-price/verified-accumulator-x3/v2 describes a workflow that verifies an external price message, constructs authenticated price information, reuses it across three feed updates, and then closes the temporary container. It called the same two package revisions on all 21,016 instances: 0x55300367a2d40813727ccac4ecee977a39fb9cdb46f2e6b2c354b9798f5de2c0 and 0x99de5c967d8206ef4b75c0afab3df2a59eb02b05c282821db803831008ac25b4.
Transaction 3JEpDDFm is an early example. Its narrative records verification, authenticated price-information construction, a coin split, and three price-related calls. The latest transaction, D2gcTCLQ, has the same seven-step shape and the same reported sender movement: −0.000352052 SUI. The detail explicitly cautions that running the sequence does not establish price freshness or correctness, and the evidence does not resolve whether each individual update changed stored price data.
The capped latest-500 stream covers only the final 14 minutes and 53 seconds of this address's activity, from 23:22:44 to 23:37:38 UTC. All 500 rows in that stream assert update-price; none failed. That recent slice therefore mirrors the complete-history character rather than revealing a new mode.
The exception
The one behavioral deviation in the complete history was HpGQ71T9, recorded at 18:39:11 UTC. It used a two-step shape: split native value, then transfer the resulting objects. Its narrative describes custody transfer to a resolved recipient, while warning that a recipient who is also the sender would not establish a custody change. The available evidence therefore supports “a native-asset split followed by object transfer,” not a claim that this was a payment or payout.
What moved
Across the complete history, the sender's SUI movements show 7.695243712 SUI out and 0.29321624 SUI in, for a net of −7.402027472 SUI in the gas evidence category. The same figures appear in the holding evidence category because the gas charge is also represented as a holding balance change; those categories must not be added together. No priced cross-asset value or profitability conclusion is available.
The address paid its own gas in every stored transaction. There is no sponsorship signal here, so the gas owner and sender are the same on-chain identity. That still says nothing about the person, service, or organization behind the address.
To continue, inspect D2gcTCLQ for the latest repeated workflow, HpGQ71T9 for the lone transfer deviation, or follow sui/update-price/verified-accumulator-x3/v2 across the two package revisions above.