Raycaster / evalsBack to SpreadsheetBench

SpreadsheetBench

105-24

1/1Pass

SpreadsheetBench task 105-24. Inspect the exact spreadsheet prompt, compare published model runs, open the agent response trace, and review grades for workbook editing.

Spreadsheet editingDual harness
ssb-105-24
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): Open!'B1:P23,'Productivity!'A1:E21' Expected answer region(s): 'Open'!C2:C23 Expected answer sheet(s): Open Task: In my Excel file, the 'Open' sheet contains IDs in Columns C and K. Find all these IDs and replace them with their respective Names. The corresponding Names for these IDs are located in Columns A and B of the 'Productivity' sheet. I require assistance with creating a macro or to automate this replacement process and also automatically consider any new additions to Columns A and B in the future for replacements. Do not create instructions, explanations, or a ID_Lookup_Helper tab. Do not create add data to any other columns other than C or K in the Open tab. Do not hide any data.

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