Raycaster / evalsBack to SpreadsheetBench

SpreadsheetBench

374-18

1/1Pass

SpreadsheetBench task 374-18. Inspect the exact spreadsheet prompt, compare published model runs, open the agent response trace, and review grades for workbook editing.

Spreadsheet editingDual harness
ssb-374-18
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): A1:G6 Expected answer region(s): 'Sheet1'!A2:G6 Expected answer sheet(s): Sheet1 Task: I attempted to write a macro to delete all rows in Column E of the 'Imported Data' sheet that have values less than 1, but I received a compile error ("End If without Block If"). Please correct and test the code so it runs successfully without adding or removing any sheets or columns, and ensure it efficiently deletes all rows in Column E where the value is < 1 while keeping the header intact. The macro should execute directly on the 'Imported Data' sheet.

Published trajectories

Agent runs on this task

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

ModelHarnessScoreResultLinks
GPT-5.4showcasedual1/1Pass