Q1 occupancy per cabin per month
The request
The task request supplied to the adapter, alongside the workspace below. Runtime instructions, authorized tools and skills also affect the system. Containerized runs withhold host-side references; local mode is not an isolation boundary.
I need Q1 occupancy for each of our cabins for the owners' letters - one row per cabin, January to March across. Save it as occupancy.xlsx with live formulas. Ruth's note says how the owner agreements count it.
Workspace
- blocked_dates_all_cabins.ics
- occupancy_note_from_ruth.txt
- reservations_export_2026-04-02.csv
Expected deliverables
- occupancy.xlsx
This task also has a task-specific grader (check.py).
Recorded acceptance checks
The table uses check names and outcomes from the frozen campaign ledger. Acceptance requires all required predicates to pass; a predicate may itself contain a tolerance or coverage threshold. Its scope is part of the measurement contract.
The frozen scorer's check set differs from this task's source declaration. These are different evaluator versions. The counts below report only checks actually recorded; a missing check is never counted as a pass.
| Recorded check | Type | Required | Failed / observed |
|---|---|---|---|
| occupancy.xlsx exists | file_exists | yes | 0 / 6 |
| live formulas | xlsx_has_formulas | yes | 0 / 6 |
| no error cells | xlsx_no_errors | yes | 0 / 6 |
| occupancy rates per cabin per month (Fox Den Jan and Feb, Aspen Hollow Jan and Mar, Bear Creek Feb, Cedar Loft Feb, Summit A-Frame Jan) | custom | yes | 0 / 6 |
Current task declaration (8 checks)
From task.yaml. Original runner grading and frozen retrospective scoring are separate. Read the frozen source for the rules behind the reported results.
- occupancy.xlsx exists —
file_exists, required - live formulas —
xlsx_has_formulas, required - no error cells —
xlsx_no_errors, required - Aspen Hollow Q1 booked nights (stays split at month ends) —
xlsx_value_present, required - Bear Creek Lodge Q1 booked nights (no cancellations or owner stays) —
xlsx_value_present, required - Fox Den Cabin Q1 available nights (blocks, DTEND exclusive) —
xlsx_value_present, required - total Q1 booked nights —
xlsx_value_present, required - occupancy rates for Fox Den February, Aspen Hollow January and the portfolio quarter —
custom, required
Recorded results, desk release
| System | Repetitions | Frozen / 3 | Original / 3 | Median minutes | Mean model estimate (USD) |
|---|---|---|---|---|---|
| Proto / DeepSeek V4.1 Flash | 3 | 0 | 4.0 | 0.0650 | |
| Codex / GPT-5.6 sol | 3 | 0 | 4.9 | 0.5616 |
These are repeated executions on the released fixture, not independent new business cases. Cost is captured model usage priced under the recorded assumptions, excluding infrastructure and human review. Per-attempt check verdicts, raw and frozen verdicts, usage, receipt and artifact hashes for this task are in results/latest/attempts.jsonl under "task": "occupancy-report".
Recorded results, desk comparison, 28 September 2026
| System | Repetitions | Frozen / 3 | Original / 3 | Median minutes |
|---|---|---|---|---|
| Proto / GPT-6 sol | 3 | 0 | 2.7 | |
| Codex / GPT-6 sol | 3 | 0 | 3.5 | |
| Proto / DeepSeek V4.1 Flash | 3 | 0 | 2.2 |
A separate campaign under the same frozen scorer, not pooled with the release. Conditions and data.