Tasks · desk · Spreadsheet
October rent roll for the bank
rent-roll-build · 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.
Carol needs a rent roll for the bank from the RentEasy exports in this folder. Save it as rent_roll.xlsx with the total as a formula; her email says what goes on it.
Workspace
- email_from_owner.txt
- lease_export_2026-10-01.csv
- tenants.csv
- unit_list.xlsx
Expected deliverables
- rent_roll.xlsx
This task also has a task-specific grader (check.py).
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 |
|---|---|---|---|
| live formulas | xlsx_has_formulas | yes | 0 |
| no error cells | xlsx_no_errors | yes | 0 |
| total monthly rent | xlsx_value_present | yes | 0 |
| month-to-month rent with the fee | xlsx_value_present | yes | 0 |
| every unit with tenant, rent and renewal flag | custom | yes | 1 |
Recorded results, complete desk comparison
This is one of the 7 tasks with a task-specific equivalence grader in the frozen scorer, which accepts alternative valid representations the original checks rejected. Raw and frozen verdicts can therefore differ here by design; see what the frozen scorer changed.
| System | Repetitions | Frozen / 3 | Raw / 3 | Median minutes | Mean cost (USD) |
|---|---|---|---|---|---|
| Proto / DeepSeek V4.1 Flash | 3 | 2 | 1.8 | 0.0226 | |
| Codex / GPT-5.6 sol | 2 | 2 | 3.5 | 0.4640 |
Per-attempt check verdicts, raw and frozen verdicts, usage, receipt and artifact hashes for this task are in results/latest/attempts.jsonl under "task": "rent-roll-build".