Raycaster / evalsBack to SpreadsheetBench

SpreadsheetBench

37229

1/1Pass

SpreadsheetBench task 37229. Inspect the exact spreadsheet prompt, compare published model runs, open the agent response trace, and review grades for workbook editing.

Spreadsheet editingDual harness
ssb-37229
SpreadsheetBench
1 model
Dual (parsed + original)
PromptWorkbook objective and answer cells
ResponseOpen each trace to inspect tool use and edits
GradeCompare score and pass/fail by model

Task prompt

What the agent was asked to do

You are solving a spreadsheet benchmark task in a real workbook. Objective: Produce the correct final workbook state for the expected answer region. What matters: - Only the values in the expected answer region will be graded. - The workbook is the answer. Instructions: 1. Read the workbook and inspect the relevant data region first. 2. Infer the required result for the provided workbook instance. 3. Write the final value(s) directly into the expected answer region. 4. Do not rely on prose, formulas in your chat response, pseudocode, or VBA as the answer unless the benchmark explicitly requires those to be written into cells. 5. If the natural-language task asks for a general method, formula, or macro, convert that into the concrete result needed for this workbook instance. 6. Keep your final text response short and only summarize the workbook cells you changed. Relevant data region(s): Expected answer region(s): C3:G14 Expected answer sheet(s): Task: How can I create a formula in Excel that will sum numbers entered in the range A1:G24 based on month end dates entered in the range B17:B24, and display the totals in the corresponding areas of C3:G14? These total cells are associated with each month of the year presented in B3:B14, and the expected results are shown in C3:G14 to use as a reference guide for the required calculations. If the sum is zero, you can leave the cell as empty.

Published trajectories

Agent runs on this task

Curated dual-harness runs. Best scored run per model.

ModelHarnessScoreResultLinks
GPT-5.4showcasedual1/1Pass