Raycaster / evalsBack to SpreadsheetBench

SpreadsheetBench

208-20

1/1Pass

SpreadsheetBench task 208-20. Inspect the exact spreadsheet prompt, compare published model runs, open the agent response trace, and review grades for workbook editing.

Spreadsheet editingDual harness
ssb-208-20
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): A1:F37 Expected answer region(s): 'Q'!A1:F37 Expected answer sheet(s): Q Task: How can I sort column I, labeled 'R DAYS', in an Excel spreadsheet from highest to lowest using VBA or other options? The spreadsheet I am working on is not sorting correctly with the code I tried, which attempts to sort from A3 to F600 based on column E in ascending order. There are issues with the code possibly because the data isn't contiguous, meaning there might be empty fields causing the sort to fail. I have provided an Excel file showing the current results that need to be adjusted (6 columns under ‘Actual Results’ table) and the desired outcome results (6 columns under ‘Expected Results’ table) for reference. Reorder the ‘Actual Results’ table to match ‘Expected,’ while leaving all formatting and values as-is.

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