APEX-Agents · Law
World425_tas_04
APEX-Agents task World425_tas_04 in AI Agents for Tax Due Diligence. Compare dual-harness agent runs across models, scores, and public traces.
Task prompt
What the agent was asked to do
Harbor Bridge has recently expressed concern that certain provisions of Summit's Amended and Restated Shareholder Agreement may be inconsistent with Summit's S-Corp election under 26 U.S.C. § 1361 and 26 C.F.R. § 1.1361-1(l). Please examine the shareholder agreement and draft a concise amendment to the agreement to correct any potential deficiencies which may invalidate Summit's S-Corp election. Put everything in a new docx file.
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 Flash | dual | 3/4 | Fail | Run detailsPublic trace |
| GPT-5.4 | dual | 3/4 | Fail | Run detailsPublic trace |
| GPT-5.5 | dual | 3/4 | Fail | Run detailsPublic trace |
| fireworks models Kimi K2 | dual | 2/4 | Fail | Run detailsPublic trace |
| Gemini 3.1 Pro | dual | 2/4 | Fail | Run detailsPublic trace |
| GPT-5.4 mini | dual | 2/4 | Fail | Run detailsPublic trace |
| GPT-5.4 nano | dual | 2/4 | Fail | Run detailsPublic trace |
Grading rubric
Rubric criteria
Runs are graded against these criteria. Open a run for model-specific verdicts.
States that amended Section 2.3 to require strictly pro rata distributions, in accordance with each Shareholder’s percentage ownership of the Company’s stock
States that Section 2.5 is deleted in its entirety
States that except as specifically amended, the original agreement shall remain in full force and effect
States that the amendment will be governed and construed by the laws of the State of Delaware