APEX-Agents · Law
World421_AP_02
APEX-Agents task World421_AP_02 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 received a complaint from Barbara Smith who alleges that SLL sent eleven text messages to a cell phone number that she had placed on the National Do-Not-Call Registry and provided to her eighty-five year old mother. Her mother opted-in to receive automated promotional messages through a webform after clicking on an advertisement. Can you review all analogous cases to the issue and determine whether SLL can be held liable for sending these text messages? Please briefly explain your conclusion on SLL’s liability in one or two sentences. I would also like you to give me the three most applicable cases and their holdings. Write your answer to me 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 mini | dual | 8/9 | Fail | Run detailsPublic trace |
| fireworks models Kimi K2 | dual | 6/9 | Fail | Run detailsPublic trace |
| GPT-5.4 | dual | 5/9 | Fail | Run detailsPublic trace |
| GPT-5.4 nano | dual | 5/9 | Fail | Run detailsPublic trace |
| GPT-5.5 | dual | 5/9 | Fail | Run detailsPublic trace |
| Gemini 3.1 Pro | dual | 1/9 | Fail | Run detailsPublic trace |
Grading rubric
Rubric criteria
Runs are graded against these criteria. Open a run for model-specific verdicts.
States that one of the analogous cases that is most applicable to the facts is Hall v. Smosh Dot Com Inc
States that one of the analogous cases that is most applicable to the facts is Wilson v. Skopos Financial, LLC
States that one of the analogous cases that is most applicable to the facts is Satterfield v. Simon Schuster Inc
States that a telemarketing text sent to a number listed on the Do-Not-Call Registry violates the TCPA, even if the communications are intended for (or solicited by) another individual
States that a text message falls under the scope of a "call" for the purposes of the TCPA
States that communication to phone numbers via text message is covered by the National Do-Not-Call Registry
States that SLL sent text messages to Barbara Smith's cell phone number without her consent
States that SLL sent text messages to Barbara Smith's cell phone number with the knowledge that the number was on the National Do-Not-Call Registry
States that SLL is likely to be held liable for a TCPA violation