APEX-Agents · Law
World 415-CD-02
APEX-Agents task World 415-CD-02 in AI Agents for SEC Disclosure Analysis. Compare dual-harness agent runs across models, scores, and public traces.
Task prompt
What the agent was asked to do
Review the following: the statement dated July 19, 2024 from George Kurtz, the Form 8-K from Crowdstrike dated July 19, 2024 and Rule 10b-5. Based solely on these documents, identify any statements made by Kurtz that are clearly and unequivocally misleading under Rule 10b-5. For any problematic statement(s), identify the missing context that is required for them. Reply back to me with your findings as an answer here
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 | dual | 2/2 | Pass | Run detailsPublic trace |
| GPT-5.4 mini | dual | 2/2 | Pass | Run detailsPublic trace |
| GPT-5.4 nano | dual | 2/2 | Pass | Run detailsPublic trace |
| GPT-5.5 | dual | 2/2 | Pass | Run detailsPublic trace |
| fireworks models Kimi K2 | dual | 0/2 | Fail | Run detailsPublic trace |
| Gemini 3 Flash | dual | 0/2 | Fail | Run detailsPublic trace |
| Gemini 3.1 Pro | dual | 0/2 | Fail | Run detailsPublic trace |
Grading rubric
Rubric criteria
Runs are graded against these criteria. Open a run for model-specific verdicts.
States that nothing in Kurtz's statement materially contradicts the disclosure provided in the Form 8-K
States Kurtz statement is not misleading under Rule 10b-5