{
  "status": "passed",
  "identity": {
    "backend": "nemo_gym",
    "protocol": "hessian-gym-v1",
    "upstream_revision": "720b5f318caee339bfd29c554c59a97edcedc846",
    "upstream_version": "0.7.0rc0",
    "upstream_lock_sha256": "1f9539274322df5b3b869addfdfda40a7c3c639ed21a46ddee2ccecddb81f136",
    "environment_sha256": "b62f23111b55f1c12d48364521b5799e19b6c716514200a79db1b8d553b13bda",
    "server_sha256": "8cc9ad8efc50cea8af2ec7f4fce868e73bf449fc6c820a530281a15bce14dcf5",
    "python": "3.13.14",
    "framework_class": "nemo_gym.base_resources_server.SimpleResourcesServer",
    "inference": false,
    "max_sessions": 256
  },
  "checks": [
    "32_concurrent_isolated_real_client_episodes_verify_and_close",
    "environment_reset_reuse_beyond_server_capacity_has_no_active_session_leak",
    "256_active_session_limit_fails_closed_then_recovers_after_cleanup",
    "actual_TRL_unknown_name_kwargs_and_unexecuted_cap_diagnostics",
    "real_service_restart_loses_old_session_stickily_and_allows_new_independent_episode"
  ],
  "check_count": 5,
  "details": {
    "sequential_reused_environment_episodes": 260,
    "native_dispatch": {
      "status": "passed",
      "trl": "0.29.0",
      "upstream_trainer_sha256": "7a112604cd51186c73817b4d00ac9e2d752d674b8e013581395b27732d9321c7",
      "cases": [
        {
          "case": "unknown_name",
          "declared_calls": 3,
          "actual_rpc_calls": 2,
          "tool_result_messages": 3,
          "native_dispatch_failures": 1,
          "native_only_error_results": 1,
          "strict_success": false,
          "trusted_tool_errors": 0,
          "partial_training_reward": 0.4
        },
        {
          "case": "unexpected_keyword",
          "declared_calls": 3,
          "actual_rpc_calls": 2,
          "tool_result_messages": 3,
          "native_dispatch_failures": 1,
          "native_only_error_results": 1,
          "strict_success": false,
          "trusted_tool_errors": 0,
          "partial_training_reward": 0.4
        },
        {
          "case": "unexecuted_at_cap",
          "declared_calls": 3,
          "actual_rpc_calls": 2,
          "tool_result_messages": 2,
          "native_dispatch_failures": 0,
          "native_only_error_results": 0,
          "strict_success": false,
          "trusted_tool_errors": 0,
          "partial_training_reward": 0.4
        }
      ],
      "model_inference": false,
      "limitation": "Actual upstream tool-loop Python dispatch with a scripted byte processor; no model, optimizer, or sampled-token qualification."
    }
  },
  "completed_at": "2026-10-20T09:53:28.534037+00:00",
  "wall_seconds": 22.529678497463465,
  "source_sha256": {
    "examples/agent_gyms/nemo_gym_stress.py": "12af94d8d89913370edd6df3f94640fc47249c270b0a12f0d32381ca3fefcd39",
    "examples/agent_gyms/nemo_gym_server.py": "8cc9ad8efc50cea8af2ec7f4fce868e73bf449fc6c820a530281a15bce14dcf5",
    "examples/agent_gyms/nemo_gym_client_smoke.py": "43fe727a055083ec24e998c9826a14917a6541a630c17162031bdcbdca8d8613",
    "src/hessian_recipes/gym_client.py": "6371e637cdb80cfad5a56f3670079e8c24abbc9b10631ce3d13a3f228fca5166",
    "src/hessian_recipes/agent_rl_tasks.py": "b62f23111b55f1c12d48364521b5799e19b6c716514200a79db1b8d553b13bda",
    "src/hessian_recipes/agent_rl_training.py": "24a0f1e132f1d941607824a11ceae8009bdb4928feda53a4ba4b065e54529ae5"
  },
  "limitations": "CPU infrastructure and adversarial protocol qualification using scripted TRAIN fixtures. No GPU allocation, model inference, learning, heldout score, fluency or general-agent claim.",
  "raw_report_sha256": "0aab00a5e9370e825a82f610f1d3623ff687c65de2d1af45a1c3df6203b9fdce",
  "raw_report_location": "runs/agent-gyms/nemo-stress-20261020-02/report.json"
}