Raycaster / evalsBack to SpreadsheetBench

SpreadsheetBench

302-1

1/1Pass

SpreadsheetBench task 302-1. Inspect the exact spreadsheet prompt, compare published model runs, open the agent response trace, and review grades for workbook editing.

Spreadsheet editingDual harness
ssb-302-1
SpreadsheetBench
2 models
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): ITEMS!'A1:B8,'SHEET1!'A1:B8,'SHEET2!'A1:B7,'result sheet1!'A1:B8,'result sheet2!'A1:B7' Expected answer region(s): 'SHEET1'!B2:B8,'SHEET2'!B2:B7 Expected answer sheet(s): 'Sheet1','Sheet2' Task: How can I match and replace the values in Sheet1 and Sheet2 with those in the ITEMS sheet based on column B? Specifically, if the value in cell B2 of Sheet1 or Sheet2 does not match the value in cell B2 of the ITEMS sheet, then B2 in Sheet1 or Sheet2 should be replaced with the corresponding value from the ITEMS sheet. This comparison and replacement should be done for each cell in column B. After the replacement, column B in both Sheet1 and Sheet2 should match column B in the ITEMS sheet - the outcome I'm expecting are detailed in 'result sheet1' and 'result sheet2'. My actual dataset could be up to 500 rows.

Published trajectories

Agent runs on this task

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

ModelHarnessScoreResultLinks
GPT-5.4showcasedual1/1Pass
Gemini 3.1 Produal1/1Pass