This on-chain address resembles a collection of vault-custodied route attempts more than a clean trading record. Across 80 transactions between 13:25:09 and 23:35:45 UTC, 47 failed and 33 did not carry an execution error. The evidence does not establish who operated it, whether the routes were profitable, or whether any exchange completed.
The aggregate figures here cover all 80 transactions in the frozen window. The row-level ledger and prose stream is capped at the latest 500 transactions; this address has only 80, so the detailed stream contains the entire observed set.
A stop-start rhythm with repeated failure points
The median gap was 277.827 seconds, with a 90th-percentile gap of 1,075.128 seconds and a longest idle period of about 49 minutes 36 seconds. The busiest minute contained two transactions. The longest consecutive run of failures was eight, matched by a longest run of eight transactions without an execution error.
Failure was not evenly distributed. Forty-four of the 47 failures stopped at 0xf0b2cc…c9d5c6::vault::deposit_with_threshold_and_event, accounting for 93.6% of failures. Two stopped at an unrecorded site, and one at 0xf6a6d2…5c4b1b::price_oracle::get_valid_price. An aborted Sui transaction rolls back application effects, so reaching a swap or deposit call proves an attempt, not a durable result.
Mostly routes, often without a named interpretation
Only 22 of the 80 rows had an asserted execute-swap-route behavior. The remaining 58 were described through their connected calls without a whole-transaction behavior claim. The named route variants were mostly one-offs: the five most frequent variants appeared twice each, while the others appeared once.
The authored route descriptions repeatedly frame the activity as a vault withdrawal followed by adapter, pool, router, or flash-swap steps and a thresholded return to the vault. That is a description of the composed mechanism, not proof that an exchange occurred. The route inputs and minimum-output guards are not receipts.
For example, 3i1YQBpr is described as withdrawing vault-held value, passing through an adapter and flash-swap lifecycle, and returning toward the vault. It submitted 10 USDC and guarded on 10 USDC, but the transaction detail explicitly says execution does not establish an exchange. Its sender-bound movement was only −0.00050942 SUI.
The failed GBzwrJnv followed a pool-buy and return-swap composition using 403.571428571 SUI, then aborted at the vault deposit step. The stored detail says earlier application changes were rolled back. The latest row, 9JurPKAd, used a different SUI → STSUI → USDC → NS route before reaching an unexplained adapter call and aborting at the same deposit entrypoint.
The largest gas charge was 956TVv9q: 8,849,636 MIST, or 0.008849636 SUI. It touched several exchange-related calls, including LBTC, USDC, and SUI, but its own detail still leaves the exchange outcome unresolved.
The bounded money story
Across all 80 transactions, the stored sender movement evidence records 0.06258325 SUI out as gas. The same amount also appears as a sender holding delta; those are two evidence views of the same outflow and must not be added together. No other sender asset movement is recorded in this scope, and the gas owner matched the sender on every transaction.
That supports a precise conclusion: this address spent 0.06258325 SUI in recorded gas during the window. It does not support a profit, loss, exchange volume, or completed vault-return claim. The submitted amounts and minimum-output guards describe attempted route parameters, not what the sender ultimately received.
Useful exits are 3i1YQBpr for a committed flash-swap-shaped route, GBzwrJnv for a route that rolled back at the vault deposit, and the recurring 0xf0b2cc…c9d5c6::vault::deposit_with_threshold_and_event call where most failures stopped.