Execution audit / 01


The last export we read said $205,901 of volume. $165,964 of it was real.

Use our export and see where the other $39,937 went
Analyse a file or use our export

Observability for the bots that trade for you — Hummingbot, Freqtrade, or your own code. They move real money and run with almost no instrumentation. Drop in a fill history export and Vexxsus takes the result apart: fees, maker/taker premium, fragmentation, timing, and the price spread inside a single order.

Execution map

One live account · 1,020 fills from 1,112 rows · 5 weeks

Volume traded

$165,964

Fees paid

$128.79

Avoidable per $1M traded

$412.20

Maker share

35.8%

Orders fragmenting

46%

Worst order

23 fills

Fees avoidable

53%

Run it / 02


The engine ships with the page and runs in this tab. Open the network panel, drop a file, watch the counter stay at zero. A fill history never leaves the machine it is on, because there is nowhere for it to go.

The figures at the top of this page are real.Measured on one live account with the owner's permission: 1,020 fills, 38 instruments, five weeks, $165,964 traded on Bybit perps. Nothing there is modelled. The sample button below is the exception and says so — it generates an export so you can see the report without having a file to hand.

Drop a fill history export

Reads Bybit, OKX and Binance fill exports (API JSON or the trade-history CSV), Hummingbot and Freqtrade. Anything else goes through a generic CSV reader. Verified on real API responses from OKX and Binance, and one real Bybit export. The OKX trade-history CSV is read from OKX's published column list and has not been checked against a real export yet. No API key, no account.

Off-origin requests0

What it measures / 03


Measure

Read every fill and price it against your own maker rate.

Attribute

Split the cost by instrument, by hour, by strategy.

Act

A ranked list of what to fix, ordered by what it costs you.

What each input answers / 04


A fill export records what you paid, never what the book looked like when you paid it. The agent closes that gap from your own machine, with a minute of public candles from the venue you already trade on — then runs on a schedule, so today's number stands against your own history and a drift shows up as a drift.

What the file answers, and what the agent adds

Why nothing exists / 05


Institutional TCA is sold on annual contracts to funds with compliance departments. Trade journals are built for a person clicking buy a few times a day. Between them sit the desks and firms running bots at size: too small for the first, too serious for the second, and nobody has built for them.

Who this is for, and why the gap is there

What we can see / 06


Every tool that touches trading data promises to be careful. A promise is worth nothing, so this is architecture: fills are never sent, history is sealed before it is stored, and the few numbers that do leave cannot be traced to an account.

0

fills that reach us, at any tier

8

rounded numbers, the whole payload

20

accounts before a cohort returns

The three layers, and the payload printed in full

Pricing / 07


The first audit is free, because seeing the number once is the whole argument. Paid access opens on request.

$0

free — one audit, no account

On request

paid access, scoped to the job

The free audit, and how to ask about paid access

Ask about the agent

The agent is the paid monitor: it reads your exports on a schedule and prices your fills against the market, from your machine. It is built; access opens on request, scoped to what you run. Leave an email or a Telegram handle and we will come back to you.

Stored on our own server, never passed to anyone, deleted on request. No fill data is attached to this.

Find out what yours costs.

A fill history export from any venue. Read, analysed and rendered in this tab — about a second for a few thousand fills, and still nothing uploaded.

Analyse a file