THE DOSSIER · one of 50

A route assembler that kept changing shape

0xc6037a53638b403b9b988f86cb3054d8087480d23238eb4d84da3481885a39e8
68 transactions in the window · 48 stopped early · browse every one →

For 10 hours and 27 minutes, this address assembled 64 different command-and-call shapes around vault withdrawals, swaps, and deposits. Of 68 eligible transactions, 48 failed; 46 stopped at the same vault::deposit_with_threshold_and_event call. The activity resembles repeated automated route attempts, but the evidence does not establish who operated the address or whether any route was profitable.

Repetition without a fixed script

Whole-history totals cover all 68 stored transactions. Of the 67 transactions with a usable transaction-level classification, 17 asserted execute-swap-route; the remaining 50 received only a composition-level explanation. Those 17 route assertions used 16 pattern variants, so even the semantically recognized subset was not one repeated script.

The cadence was sparse and irregular rather than continuous: the median gap was 396,292 milliseconds, about 6 minutes 36 seconds; the 90th-percentile gap was about 20 minutes 27 seconds; and the longest idle period was roughly 48 minutes. The busiest minute contained only two transactions, at 15:11 UTC.

The route family itself is consistent. The authored explanations describe withdrawals from vault custody, intermediate adapter or pool exchanges, coin-and-balance conversions, and a thresholded deposit back toward the vault. They also make the important limitation explicit: a submitted amount or minimum-output guard is not proof of spent value or received output.

The failure boundary

Across the complete eligible history, 48 transactions failed. Forty-six stopped at the vault deposit boundary and two have no recorded abort site. On the 67 transactions with usable classifications, 47 failed. The longest consecutive failure run was 11 transactions.

A representative successful execution, 7ads3HvG, was structurally understood as a seven-call route. It submitted 14.5 SUI and guarded on 14.5 SUI as the minimum output, but its own explanation says that execution does not establish an exchange. The sender-level movement recorded for the transaction was only −0.000386292 SUI, shown separately as holding and gas evidence.

3QVgY6hK followed another seven-call vault-funded route with a 10.95 SUI guard, then aborted at the same deposit function. Because Sui rolls back application effects after an abort, the reached withdrawal and exchange-related calls prove an attempted path, not a completed custody return or exchange.

The latest observed transaction, DAsk53d9, is an instructive exception to the otherwise purely negative SUI movement record. Its eight connected calls included a vault withdrawal, several exchange-related steps, and a deposit attempt. The sender boundary showed −0.006615912 SUI and +55,487 units of the asset displayed as 0x514549…NS. The route explanation still says that the exchange outcome is unresolved.

What actually moved

The complete-history value evidence is narrow:

  • Gas evidence records −74,135,966 MIST, or −0.074135966 SUI, across all 68 transactions.
  • The holding evidence separately records −74,135,968 MIST, or −0.074135968 SUI, also across 68 transactions. These lanes are not additive; the same gas charge can appear in both.
  • Holding evidence records +55,487 units of 0x514549…NS in one transaction. No decimal precision or price is provided, so this cannot be converted into a value comparison or profit claim.

The largest recorded gas charge was 14,664,288 MIST in 9stvFv71; that transaction also stopped at the recurring vault deposit boundary. The evidence therefore supports a story of costly attempted routing, not a measured exchange result.

Scope and identity

The detailed transaction stream is capped at 500 rows, but this address has only 67 transactions with usable classifications, so the ledger and detail views cover all of them. The address was never sponsored in this window: all 68 transactions used the address itself as gas owner. That establishes a funding relationship for these rows, not a person, company, operator, or bot identity.

The clearest description is therefore behavioral: a sparse, shape-diverse address repeatedly assembled vault-centered swap routes, often reaching the same terminal deposit guard and usually aborting there. What the routes were intended to accomplish, and whether any value beyond the recorded boundary movements changed hands, remains unresolved.

Dossiers are AI-authored readings of one address's stored window. They describe observed behavior — never operators, identity, or profit the evidence doesn't establish.

everything this address did → all dossiers the shelf