Tasks · desk · Bookkeeping
Realized FX gain or loss on the euro invoices
fx-invoice-gains · timeout 1200 s · source
The ask
Sent verbatim as the prompt. The agent works in a copy of the workspace below and sees nothing else.
The auditors want the FX gain or loss on each of our euro invoices from the first half. The invoice register, the bank's wire report and the ECB rates are in the folder, and Miriam's note says how she books it. Save it as fx_results.csv.
Workspace
- ar_invoice_register_2026-H1.xlsx
- ecb_reference_rate_usd_2026.csv
- harbor_trust_incoming_wires_2026-01-01_to_2026-08-31.csv
- note_from_miriam_controller.txt
Expected deliverables
- fx_results.csv
Checks
The task passes only when every required check passes. Reference answers, tolerances, and the planted traps are in the repository, not here.
| Check | Type | Required | Failed (frozen) / 6 |
|---|---|---|---|
| requested columns | csv_columns | yes | 2 |
| which invoices | csv_set_equal | yes | 0 |
| row count | csv_row_count | yes | 0 |
| fx gain or loss per invoice | csv_values_match | yes | 2 |
| euros still open | csv_values_match | yes | 0 |
Recorded results, complete desk comparison
pass pass, frozen scorer only fail timed out
| System | Repetitions | Frozen / 3 | Raw / 3 | Median minutes | Mean cost (USD) |
|---|---|---|---|---|---|
| Proto / DeepSeek V4.1 Flash | 3 | 3 | 1.2 | 0.0181 | |
| Codex / GPT-5.6 sol | 1 | 1 | 1.5 | 0.2341 |
Per-attempt check verdicts, raw and frozen verdicts, usage, receipt and artifact hashes for this task are in results/latest/attempts.jsonl under "task": "fx-invoice-gains".