Raycaster / evalsBack to SpreadsheetBench

SpreadsheetBench

267-21

0/1Fail

SpreadsheetBench task 267-21. Inspect the exact spreadsheet prompt, compare published model runs, open the agent response trace, and review grades for workbook editing.

Spreadsheet editingDual harness
ssb-267-21
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): SH1!A1:C8','RP1!A1:C8','merging!A1:D11' Expected answer region(s): merging'!C2:D11 Expected answer sheet(s): 'SH1','RP1','merging' Task: How can I create a comprehensive report in a single sheet ('sheet merging') in Excel that compiles data from multiple sheets, with each sheet representing monthly data that will be added every month, and ensure that the merged report sheet updates as new data is added in the subsequent monthly sheets? Please output the answer in the corresponding positions of columns C and D in the table, and pay attention to the mapping between the answer and column name and row value. Basically, look up ID from Column B and place the corresponding QTY value from SH1 into Column C and from RP1 into Column D. If there's an error or blank, insert "-".

Published trajectories

Agent runs on this task

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

ModelHarnessScoreResultLinks
GPT-5.4showcasedual0/1Fail
Gemini 3.1 Produal1/1Pass