Spaces:
Running
Running
| { | |
| "domain": "tau2", | |
| "scenario_id": "TicketTwister", | |
| "mission_id": "tickettwister_tau2_sample", | |
| "reset_payload": { | |
| "seed": 42, | |
| "scenario_id": "TicketTwister", | |
| "mission_id": "tickettwister_tau2_sample", | |
| "options": { | |
| "env_id": "omnibench_aegis_env:tau2.tickettwister", | |
| "max_steps": 5, | |
| "target_score": 1, | |
| "domain": "tau2" | |
| } | |
| }, | |
| "action_plan": [ | |
| { | |
| "name": "load_tasks", | |
| "args": {} | |
| }, | |
| { | |
| "name": "prepare_user", | |
| "args": {} | |
| }, | |
| { | |
| "name": "run_conversation", | |
| "args": {} | |
| }, | |
| { | |
| "name": "score_task_bundle", | |
| "args": { | |
| "task_rewards": { | |
| "tt_air_1": 1.0, | |
| "tt_air_2": 1.0 | |
| } | |
| } | |
| }, | |
| { | |
| "name": "submit_assessment", | |
| "args": {} | |
| } | |
| ], | |
| "notes": [ | |
| "Aligned fixture for the real TicketTwister tau2 domain.", | |
| "Uses the task-loading and conversation flow before scoring the bundle.", | |
| "Avoids the fallback behavior that left the run in_progress." | |
| ] | |
| } | |