APEX-Agents · Law

World417_TG_01

Best published6/9Fail

APEX-Agents task World417_TG_01 in AI Agents for Employment Law Analysis. Compare dual-harness agent runs across models, scores, and public traces.

AI Agents for Employment Law AnalysisLaw World 417Dual harnessGrader: rubric

Task IDtask_d8119ffe94d4444dbfb6626cf3eab3a2

WorldLaw World 417

Expected outputmessage_in_console

Published runs7 models · dual config

Task prompt

What the agent was asked to do

We just received a demand letter from Isaiah’s counsel. He alleges wrongful termination and FMLA interference. Can you look into the validity of his claims and return me back a write-up of what you find? I want you to just write your answer right 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-5.5	dual	6/9	Fail	Run details Public trace
Gemini 3 Flash	dual	3/9	Fail	Run details Public trace
GPT-5.4 mini	dual	3/9	Fail	Run details Public trace
Gemini 3.1 Pro	dual	2/9	Fail	Run details Public trace
GPT-5.4	dual	1/9	Fail	Run details Public trace
fireworks models Kimi K2	dual	0/9	Fail	Run details Public trace
GPT-5.4 nano	dual	0/9	Fail	Run details Public trace

Grading rubric

Rubric criteria

Runs are graded against these criteria. Open a run for model-specific verdicts.

States that Brown's employment is at-will under California law
States that illegal termination is an exception to at-will employment in California
States that violation of public policy termination is an exception to at-will employment in California
States that contractual language that prohibits termination is an exception to at-will employment in California
States that violation of the implied covenant of good faith termination is an exception to at-will employment in California
States that Brown's allegations do not fall within an exception to at-will employment under California law
States that Brown does not have a valid wrongful termination claim
States that Brown is not an eligible employee within the meaning of the FMLA
States that Chasing Streams did not violate the FMLA by denying Brown's leave request