APEX-Agents · Management Consulting
World126_JD_04
APEX-Agents task World126_JD_04 in AI Agents for ESG and Climate Risk Analysis. Compare dual-harness agent runs across models, scores, and public traces.
Task prompt
What the agent was asked to do
Please check how KO and MDLZ differ in expected upside once we apply the ESG and GLP-1 filters and account for each investor’s maximum allowable ESG risk level. Use the survey data and the ESG risk thresholds to determine which respondents are eligible to hold each company. Then calculate the confidence weighted average and standard deviation of expected annual return for KO and MDLZ. Show each company’s weighted average and standard deviation of expected annual returns. Round only the final results, going 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 |
|---|---|---|---|---|
| GPT-5.4 mini | dual | 1/4 | Fail | Run detailsPublic trace |
| Gemini 3 Flash | dual | 0/4 | Fail | Run detailsPublic trace |
| Gemini 3.1 Pro | dual | 0/4 | Fail | Run detailsPublic trace |
| GPT-5.4 | dual | 0/4 | Fail | Run detailsPublic trace |
| GPT-5.4 nano | dual | 0/4 | Fail | Run detailsPublic trace |
| GPT-5.5 | dual | 0/4 | Fail | Run detailsPublic trace |
Grading rubric
Rubric criteria
Runs are graded against these criteria. Open a run for model-specific verdicts.
States that the Weighted Avg Expected Annual Return for KO is 6.31
States that the Weighted Avg Expected Annual Return for MDLZ is 6.71
States that the Weighted Stdev Expected Annual Return for KO is 2.59
States that the Weighted Stdev Expected Annual Return for MDLZ is 2.77