APEX-Agents · Management Consulting
Task_128_PJ_01
APEX-Agents task Task_128_PJ_01 in AI Agents for Cross-Border Regulatory Review. Compare dual-harness agent runs across models, scores, and public traces.
Task prompt
What the agent was asked to do
We have conducted surveys across 3 emerging markets - EM 1, EM 2, and EM 3 to establish market potential for solar systems. 1) Based on the survey-related documents for EM1, report the percentage of small businesses that don't use solar systems in EM1 2) There were errors in the responses we received for EM1. The agency has shared an updated responses sheet that captures the correct responses for users where there was an error. Keep the original information in case the cell is blank. Also, there might be new users in this sheet. Add those users to the original file. What percentage of small businesses don't use solar systems in EM1 after accounting for this new information? Return your answers as a message here. Round all percentage values to two decimal places.
Published trajectories
Agent runs on this task
Curated dual-harness runs (parsed + original sandbox). Best scored run per model.
| Model | Harness | Score | Result | Links |
|---|---|---|---|---|
| Gemini 3.1 Pro | dual | 1/2 | Fail | Run detailsPublic trace |
| GPT-5.4 | dual | 1/2 | Fail | Run detailsPublic trace |
| GPT-5.4 mini | dual | 1/2 | Fail | Run detailsPublic trace |
| GPT-5.4 nano | dual | 1/2 | Fail | Run detailsPublic trace |
| GPT-5.5 | dual | 1/2 | Fail | Run detailsPublic trace |
Grading rubric
Rubric criteria
Runs are graded against these criteria. Open a run for model-specific verdicts.
States that in the original responses, the proportion of small businesses not using solar in EM1 is 59.11%
States that in the revised responses, the proportion of small businesses not using solar in EM1 is 59.86%