{
  "mode": "offline-synthetic-candidate-budget",
  "baseline_commit": "30524752a83438decb431b36267ac81378f0bcb8",
  "candidate": "working tree at invocation; published with its source commit",
  "baseline_source_sha256": {
    "src/apixly_jev_harness/__init__.py": "2a66fe04dc2bf2d57f18f540af469e2b5978f0eac0f78d739471cc47013974b6",
    "src/apixly_jev_harness/backend.py": "f7709b468d3d7c131fb8be42b882e5a0387db0077d1fc07e420153b0447f6754",
    "src/apixly_jev_harness/callbacks.py": "e14fa70328634bf3eae87c027d9efc33bfde3a3caebe77b388537335c7250298",
    "src/apixly_jev_harness/cli.py": "8d9434bf8a8b539163194a84e3f099761d08ef32201f99cbce46c139f3726d94",
    "src/apixly_jev_harness/context.py": "b9d76abb33f8a1ebe8731fe6d5fe6811445132a6046448fc0e2cc871507d50ea",
    "src/apixly_jev_harness/diagnostics.py": "9e8e98dc4b103c1a6687dbd31bd2dd01739309c7e4021b9ecb0f28b6976a0f71",
    "src/apixly_jev_harness/engine.py": "488efddf4ade7ba4ab47f959b94c607a40046f7f173b7fe7ac6e5826e38ba049",
    "src/apixly_jev_harness/output.py": "94be8b482e4279e29aca128a35caf54744dbe8db8188f14905488f8ddde93a0b",
    "src/apixly_jev_harness/surfaces.py": "65ce73edfe9f6a3c16de373b3a1b85671c3d2cc661f3a50dd906a15721c7d8f1",
    "src/apixly_jev_harness/task.py": "30b9dc766df860f9897032a1f902e4e1a7afbd6eff8c381d85ea9e1013a0bfe7",
    "src/apixly_jev_harness/task_cli.py": "69045fe104037bb401d20ebdd4091b1bcc01596ee7c8076cff547da69296a7e6",
    "src/apixly_jev_harness/verify.py": "cad171c63144a60d898553eb5a3d7dc0dd2189f2b20006d8edbe9f0f5a78b1e0"
  },
  "candidate_source_sha256": {
    "src/apixly_jev_harness/__init__.py": "9999dcc9843dee69b2c68b7721f8c826c89e2d6f93720e1dbae80a020f304a98",
    "src/apixly_jev_harness/backend.py": "f7709b468d3d7c131fb8be42b882e5a0387db0077d1fc07e420153b0447f6754",
    "src/apixly_jev_harness/callbacks.py": "e14fa70328634bf3eae87c027d9efc33bfde3a3caebe77b388537335c7250298",
    "src/apixly_jev_harness/cli.py": "8d9434bf8a8b539163194a84e3f099761d08ef32201f99cbce46c139f3726d94",
    "src/apixly_jev_harness/context.py": "b9d76abb33f8a1ebe8731fe6d5fe6811445132a6046448fc0e2cc871507d50ea",
    "src/apixly_jev_harness/diagnostics.py": "9e8e98dc4b103c1a6687dbd31bd2dd01739309c7e4021b9ecb0f28b6976a0f71",
    "src/apixly_jev_harness/engine.py": "1db8e15b86aa92f593b5354a1a7e8a2e168bd6e0a4c903fd37a5df336053036c",
    "src/apixly_jev_harness/output.py": "94be8b482e4279e29aca128a35caf54744dbe8db8188f14905488f8ddde93a0b",
    "src/apixly_jev_harness/surfaces.py": "65ce73edfe9f6a3c16de373b3a1b85671c3d2cc661f3a50dd906a15721c7d8f1",
    "src/apixly_jev_harness/task.py": "30b9dc766df860f9897032a1f902e4e1a7afbd6eff8c381d85ea9e1013a0bfe7",
    "src/apixly_jev_harness/task_cli.py": "d1ebbd533c59a76a99812f3692cdee80582b0c2ba832f1fa525f6ff951d93e48",
    "src/apixly_jev_harness/verify.py": "cad171c63144a60d898553eb5a3d7dc0dd2189f2b20006d8edbe9f0f5a78b1e0"
  },
  "runner_sha256": "3a73a0431ee51263da808295c54843c1c3394411e9f49d3e277e15db5eadd356",
  "python_version": "3.14.2",
  "cases": {
    "slow_false_empty": {
      "kind": "empty",
      "group": "budget_fix",
      "verify_time": 11.0
    },
    "slow_false_raises": {
      "kind": "raises",
      "group": "budget_fix",
      "verify_time": 11.0
    },
    "slow_false_task_dynamic": {
      "kind": "task",
      "group": "budget_fix",
      "verify_time": 11.0
    },
    "slow_false_generator": {
      "kind": "generator",
      "group": "budget_fix",
      "verify_time": 11.0
    },
    "prior_effect_unknown_slow_false": {
      "kind": "after_effect_empty",
      "group": "budget_fix",
      "verify_time": 11.0,
      "unknown_usage": true
    },
    "false_exact_timeout_boundary": {
      "kind": "nonempty",
      "group": "budget_fix",
      "verify_time": 10.0
    },
    "initially_expired": {
      "kind": "initial_expired",
      "group": "compatibility"
    },
    "fast_false_normal_tool": {
      "kind": "normal",
      "group": "compatibility"
    },
    "late_true_keeps_completion": {
      "kind": "late_true",
      "group": "compatibility",
      "verify_time": 11.0
    },
    "max_steps_precedes_new_guard": {
      "kind": "max_steps",
      "group": "compatibility",
      "verify_time": 11.0
    }
  },
  "compatibility_parity": true,
  "reproduction": "python benchmarks/candidate_budget.py --output local-results/candidate-budget-ab.json",
  "scope": "Ten fixed synthetic Task/Workflow/Engine fixtures, three paired repeats with alternating arm order. An explicit callback-driven monotonic clock, never sleep, models false verification reaching/passing the ten-second budget before candidate collection. Empty, raising, Task dynamic-provider and generator collectors prove collection does not begin after known exhaustion; one completed prior effect preserves steps and unknown scripted usage. Controls preserve initial expiry, ordinary fast progression, late successful verification, max-step precedence and the exact >= timeout boundary. Callback traces and journal event subsequences preserve baseline failures without raw input/body details. Selection is scripted and native inference/real HTTPTransport forbidden; actual model/network/UI calls zero if guards stay zero.",
  "negative_results": "The new guard changes terminal phase: blocked/no-candidates or a collector failure becomes budget_exceeded because that downstream callback never ran. Return bytes and event sequences therefore can increase/decrease without quality or performance meaning. Prior effects and unknown usage are retained; a late true verifier can still complete and max_steps still wins after false verification at that existing boundary. This remains a cooperative between-callback budget, not a hard deadline, callback cancellation, rollback, safe retry or a guarantee against a callback's own I/O. No AI quality, speed or fee gain is claimed.",
  "cost_conversion": "No tariff or savings assumed. Actual model usage is zero only when both guards remain zero; guard hits record unknown actual usage and fail fixtures. Scripted usage is separate from actual usage. Real perf_counter timing includes run admission, trusted callbacks, optional scripted decision/effect and private journal flush/fsync; excludes imports, synthetic clock/workflow setup and terminal analysis/source hashing. The synthetic clock only triggers budget transitions and never represents measured latency. Returned bytes omit local archive paths; scalar input/context sizes are not provider-token counts.",
  "summary": {
    "baseline": {
      "runs": 30,
      "correct": 12,
      "groups": {
        "budget_fix": {
          "runs": 18,
          "correct": 0
        },
        "compatibility": {
          "runs": 12,
          "correct": 12
        }
      },
      "statuses": {
        "blocked": 6,
        "budget_exceeded": 12,
        "done": 6,
        "failed": 3,
        "max_steps": 3
      },
      "candidate_entries": 27,
      "collector_body_calls": 27,
      "provider_calls": 3,
      "generator_yields": 3,
      "effects": 9,
      "scripted_decisions": 9,
      "unknown_scripted_usage_rows": 3,
      "native_inference_attempts": 0,
      "real_network_attempts": 0,
      "selection_input_bytes": 15156,
      "shared_context_bytes": 2448,
      "median_whole_operation_seconds": 0.0011305630032438785,
      "returned_context_bytes": 14361,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      }
    },
    "candidate": {
      "runs": 30,
      "correct": 30,
      "groups": {
        "budget_fix": {
          "runs": 18,
          "correct": 18
        },
        "compatibility": {
          "runs": 12,
          "correct": 12
        }
      },
      "statuses": {
        "budget_exceeded": 21,
        "done": 6,
        "max_steps": 3
      },
      "candidate_entries": 9,
      "collector_body_calls": 9,
      "provider_calls": 0,
      "generator_yields": 0,
      "effects": 9,
      "scripted_decisions": 9,
      "unknown_scripted_usage_rows": 3,
      "native_inference_attempts": 0,
      "real_network_attempts": 0,
      "selection_input_bytes": 15156,
      "shared_context_bytes": 2448,
      "median_whole_operation_seconds": 0.0011235834972467273,
      "returned_context_bytes": 14184,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      }
    }
  },
  "rows": [
    {
      "case": "slow_false_empty",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "empty",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "blocked",
        "complete": false,
        "reason": "no_candidates",
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.0008152089867508039,
      "returned_context_bytes": 427,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 1
    },
    {
      "case": "slow_false_empty",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "empty",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0009985000069718808,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 1
    },
    {
      "case": "slow_false_empty",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "empty",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0010801250027725473,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 2
    },
    {
      "case": "slow_false_empty",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "empty",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "blocked",
        "complete": false,
        "reason": "no_candidates",
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.0011173340026289225,
      "returned_context_bytes": 427,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 2
    },
    {
      "case": "slow_false_empty",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "empty",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "blocked",
        "complete": false,
        "reason": "no_candidates",
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.0011437920038588345,
      "returned_context_bytes": 427,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 3
    },
    {
      "case": "slow_false_empty",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "empty",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.001150333002442494,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 3
    },
    {
      "case": "slow_false_raises",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "raises",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "failed",
        "complete": false,
        "reason": null,
        "error_type": "RuntimeError",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.0009348750027129427,
      "returned_context_bytes": 429,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 1
    },
    {
      "case": "slow_false_raises",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "raises",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0009378749964525923,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 1
    },
    {
      "case": "slow_false_raises",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "raises",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0009920419979607686,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 2
    },
    {
      "case": "slow_false_raises",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "raises",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "failed",
        "complete": false,
        "reason": null,
        "error_type": "RuntimeError",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.0010394580021966249,
      "returned_context_bytes": 429,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 2
    },
    {
      "case": "slow_false_raises",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "raises",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "failed",
        "complete": false,
        "reason": null,
        "error_type": "RuntimeError",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.000995875001535751,
      "returned_context_bytes": 429,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 3
    },
    {
      "case": "slow_false_raises",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "raises",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0009977919980883598,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 3
    },
    {
      "case": "slow_false_task_dynamic",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "task",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 1,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "task_candidates",
            "synthetic_time": 11.0,
            "effects": 0
          },
          {
            "callback": "provider",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.28209479100769386,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 1
    },
    {
      "case": "slow_false_task_dynamic",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "task",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0028446659998735413,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 1
    },
    {
      "case": "slow_false_task_dynamic",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "task",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.000875749989063479,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 2
    },
    {
      "case": "slow_false_task_dynamic",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "task",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 1,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "task_candidates",
            "synthetic_time": 11.0,
            "effects": 0
          },
          {
            "callback": "provider",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.0009587920067133382,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 2
    },
    {
      "case": "slow_false_task_dynamic",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "task",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 1,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "task_candidates",
            "synthetic_time": 11.0,
            "effects": 0
          },
          {
            "callback": "provider",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.0010026669915532693,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 3
    },
    {
      "case": "slow_false_task_dynamic",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "task",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0017239999870071188,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 3
    },
    {
      "case": "slow_false_generator",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "generator",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 1,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "generator_start",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.002004541995120235,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 1
    },
    {
      "case": "slow_false_generator",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "generator",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0009980410104617476,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 1
    },
    {
      "case": "slow_false_generator",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "generator",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0010209169995505363,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 2
    },
    {
      "case": "slow_false_generator",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "generator",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 1,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "generator_start",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.0011454160121502355,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 2
    },
    {
      "case": "slow_false_generator",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "generator",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 1,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "generator_start",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.0013323750026756898,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 3
    },
    {
      "case": "slow_false_generator",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "generator",
        "group": "budget_fix",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0009895000112010166,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 3
    },
    {
      "case": "prior_effect_unknown_slow_false",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "after_effect_empty",
        "group": "budget_fix",
        "verify_time": 11.0,
        "unknown_usage": true,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": false,
        "reported_scripted_usage": null
      },
      "result": {
        "status": "blocked",
        "complete": false,
        "reason": "no_candidates",
        "error_type": null,
        "candidate_entries": 2,
        "collector_body_calls": 2,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": false,
        "reported_scripted_usage": null,
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 1,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 11.0,
            "effects": 1
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.001354291001916863,
      "returned_context_bytes": 608,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 1
    },
    {
      "case": "prior_effect_unknown_slow_false",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "after_effect_empty",
        "group": "budget_fix",
        "verify_time": 11.0,
        "unknown_usage": true,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": false,
        "reported_scripted_usage": null
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": false,
        "reported_scripted_usage": null,
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 1,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0013586249988293275,
      "returned_context_bytes": 589,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 1
    },
    {
      "case": "prior_effect_unknown_slow_false",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "after_effect_empty",
        "group": "budget_fix",
        "verify_time": 11.0,
        "unknown_usage": true,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": false,
        "reported_scripted_usage": null
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": false,
        "reported_scripted_usage": null,
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 1,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0014888330042595044,
      "returned_context_bytes": 589,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 2
    },
    {
      "case": "prior_effect_unknown_slow_false",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "after_effect_empty",
        "group": "budget_fix",
        "verify_time": 11.0,
        "unknown_usage": true,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": false,
        "reported_scripted_usage": null
      },
      "result": {
        "status": "blocked",
        "complete": false,
        "reason": "no_candidates",
        "error_type": null,
        "candidate_entries": 2,
        "collector_body_calls": 2,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": false,
        "reported_scripted_usage": null,
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 1,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 11.0,
            "effects": 1
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.0015330419992096722,
      "returned_context_bytes": 608,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 2
    },
    {
      "case": "prior_effect_unknown_slow_false",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "after_effect_empty",
        "group": "budget_fix",
        "verify_time": 11.0,
        "unknown_usage": true,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": false,
        "reported_scripted_usage": null
      },
      "result": {
        "status": "blocked",
        "complete": false,
        "reason": "no_candidates",
        "error_type": null,
        "candidate_entries": 2,
        "collector_body_calls": 2,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": false,
        "reported_scripted_usage": null,
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 1,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 11.0,
            "effects": 1
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.0014060420071473345,
      "returned_context_bytes": 608,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 3
    },
    {
      "case": "prior_effect_unknown_slow_false",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "after_effect_empty",
        "group": "budget_fix",
        "verify_time": 11.0,
        "unknown_usage": true,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": false,
        "reported_scripted_usage": null
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": false,
        "reported_scripted_usage": null,
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 1,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0015376249939436093,
      "returned_context_bytes": 589,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 3
    },
    {
      "case": "false_exact_timeout_boundary",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "nonempty",
        "group": "budget_fix",
        "verify_time": 10.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 10.0,
        "synthetic_clock_final": 10.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 10.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 10.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.0012708330032182857,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 1
    },
    {
      "case": "false_exact_timeout_boundary",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "nonempty",
        "group": "budget_fix",
        "verify_time": 10.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 10.0,
        "synthetic_clock_final": 10.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 10.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.001105165996705182,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 1
    },
    {
      "case": "false_exact_timeout_boundary",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "nonempty",
        "group": "budget_fix",
        "verify_time": 10.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 10.0,
        "synthetic_clock_final": 10.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 10.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0009560410107951611,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 2
    },
    {
      "case": "false_exact_timeout_boundary",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "nonempty",
        "group": "budget_fix",
        "verify_time": 10.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 10.0,
        "synthetic_clock_final": 10.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 10.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 10.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.0010294579988112673,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 2
    },
    {
      "case": "false_exact_timeout_boundary",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "nonempty",
        "group": "budget_fix",
        "verify_time": 10.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 10.0,
        "synthetic_clock_final": 10.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 10.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 10.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": false,
      "whole_operation_seconds": 0.000998374991468154,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 3
    },
    {
      "case": "false_exact_timeout_boundary",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "nonempty",
        "group": "budget_fix",
        "verify_time": 10.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 10.0,
        "synthetic_clock_final": 10.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 10.0,
            "effects": 0,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0012014579988317564,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 3
    },
    {
      "case": "initially_expired",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "initial_expired",
        "group": "compatibility",
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 0,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0009399580012541264,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 1
    },
    {
      "case": "initially_expired",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "initial_expired",
        "group": "compatibility",
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 0,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0011076250084443018,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 1
    },
    {
      "case": "initially_expired",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "initial_expired",
        "group": "compatibility",
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 0,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0012843330041505396,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 2
    },
    {
      "case": "initially_expired",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "initial_expired",
        "group": "compatibility",
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 0,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0010879579931497574,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 2
    },
    {
      "case": "initially_expired",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "initial_expired",
        "group": "compatibility",
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 0,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0008915409998735413,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 3
    },
    {
      "case": "initially_expired",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "initial_expired",
        "group": "compatibility",
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "budget_exceeded",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "budget_exceeded",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 0,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 11.0,
            "effects": 0
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.00092712500190828,
      "returned_context_bytes": 408,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 3
    },
    {
      "case": "fast_false_normal_tool",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "normal",
        "group": "compatibility",
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "done",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        }
      },
      "result": {
        "status": "done",
        "complete": true,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        },
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 0.0,
        "synthetic_clock_final": 0.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 1,
            "matched": true
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.001273792004212737,
      "returned_context_bytes": 644,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 1
    },
    {
      "case": "fast_false_normal_tool",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "normal",
        "group": "compatibility",
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "done",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        }
      },
      "result": {
        "status": "done",
        "complete": true,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        },
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 0.0,
        "synthetic_clock_final": 0.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 1,
            "matched": true
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.00180283299414441,
      "returned_context_bytes": 644,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 1
    },
    {
      "case": "fast_false_normal_tool",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "normal",
        "group": "compatibility",
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "done",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        }
      },
      "result": {
        "status": "done",
        "complete": true,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        },
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 0.0,
        "synthetic_clock_final": 0.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 1,
            "matched": true
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0013408750091912225,
      "returned_context_bytes": 644,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 2
    },
    {
      "case": "fast_false_normal_tool",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "normal",
        "group": "compatibility",
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "done",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        }
      },
      "result": {
        "status": "done",
        "complete": true,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        },
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 0.0,
        "synthetic_clock_final": 0.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 1,
            "matched": true
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0017391250003129244,
      "returned_context_bytes": 644,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 2
    },
    {
      "case": "fast_false_normal_tool",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "normal",
        "group": "compatibility",
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "done",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        }
      },
      "result": {
        "status": "done",
        "complete": true,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        },
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 0.0,
        "synthetic_clock_final": 0.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 1,
            "matched": true
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0017053750052582473,
      "returned_context_bytes": 644,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 3
    },
    {
      "case": "fast_false_normal_tool",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "normal",
        "group": "compatibility",
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "done",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        }
      },
      "result": {
        "status": "done",
        "complete": true,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        },
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 0.0,
        "synthetic_clock_final": 0.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 1,
            "matched": true
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0015018330013845116,
      "returned_context_bytes": 644,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 3
    },
    {
      "case": "late_true_keeps_completion",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "late_true",
        "group": "compatibility",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "done",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "done",
        "complete": true,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": true
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0011144159943796694,
      "returned_context_bytes": 396,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 1
    },
    {
      "case": "late_true_keeps_completion",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "late_true",
        "group": "compatibility",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "done",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "done",
        "complete": true,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": true
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0011131670034956187,
      "returned_context_bytes": 396,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 1
    },
    {
      "case": "late_true_keeps_completion",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "late_true",
        "group": "compatibility",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "done",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "done",
        "complete": true,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": true
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.001133999990997836,
      "returned_context_bytes": 396,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 2
    },
    {
      "case": "late_true_keeps_completion",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "late_true",
        "group": "compatibility",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "done",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "done",
        "complete": true,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": true
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.000967707994277589,
      "returned_context_bytes": 396,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 2
    },
    {
      "case": "late_true_keeps_completion",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "late_true",
        "group": "compatibility",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "done",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "done",
        "complete": true,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": true
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0011019170051440597,
      "returned_context_bytes": 396,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 3
    },
    {
      "case": "late_true_keeps_completion",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "late_true",
        "group": "compatibility",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 2,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "done",
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        }
      },
      "result": {
        "status": "done",
        "complete": true,
        "reason": null,
        "error_type": null,
        "candidate_entries": 0,
        "collector_body_calls": 0,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 0,
        "steps": 0,
        "intent_events": 0,
        "scripted_decisions": 0,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 0,
          "output_tokens": 0
        },
        "verify_calls": 1,
        "observe_calls": 1,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 0,
            "matched": true
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 0,
        "request_sizes": [],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.001011542000924237,
      "returned_context_bytes": 396,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 3
    },
    {
      "case": "max_steps_precedes_new_guard",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "max_steps",
        "group": "compatibility",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 1,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "max_steps",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        }
      },
      "result": {
        "status": "max_steps",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        },
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 1,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0016309170023305342,
      "returned_context_bytes": 651,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 1
    },
    {
      "case": "max_steps_precedes_new_guard",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "max_steps",
        "group": "compatibility",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 1,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "max_steps",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        }
      },
      "result": {
        "status": "max_steps",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        },
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 1,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.00172433300758712,
      "returned_context_bytes": 651,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 1
    },
    {
      "case": "max_steps_precedes_new_guard",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "max_steps",
        "group": "compatibility",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 1,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "max_steps",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        }
      },
      "result": {
        "status": "max_steps",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        },
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 1,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0017444580007577315,
      "returned_context_bytes": 651,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 2
    },
    {
      "case": "max_steps_precedes_new_guard",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "max_steps",
        "group": "compatibility",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 1,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "max_steps",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        }
      },
      "result": {
        "status": "max_steps",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        },
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 1,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0013176670036045834,
      "returned_context_bytes": 651,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 2
    },
    {
      "case": "max_steps_precedes_new_guard",
      "runtime_version": "0.1.15",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "max_steps",
        "group": "compatibility",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 1,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "max_steps",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        }
      },
      "result": {
        "status": "max_steps",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        },
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 1,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0015380000113509595,
      "returned_context_bytes": 651,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "baseline",
      "repeat": 3
    },
    {
      "case": "max_steps_precedes_new_guard",
      "runtime_version": "0.1.16",
      "runtime_loaded_from_requested_source": true,
      "fixture": {
        "kind": "max_steps",
        "group": "compatibility",
        "verify_time": 11.0,
        "timeout_seconds": 10.0,
        "max_steps": 1,
        "clock_advance": "explicit synthetic callback transition; no sleep"
      },
      "expected": {
        "status": "max_steps",
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        }
      },
      "result": {
        "status": "max_steps",
        "complete": false,
        "reason": null,
        "error_type": null,
        "candidate_entries": 1,
        "collector_body_calls": 1,
        "provider_calls": 0,
        "generator_yields": 0,
        "effects": 1,
        "steps": 1,
        "intent_events": 1,
        "scripted_decisions": 1,
        "usage_complete": true,
        "reported_scripted_usage": {
          "input_tokens": 7,
          "output_tokens": 2
        },
        "verify_calls": 2,
        "observe_calls": 3,
        "synthetic_elapsed_seconds": 11.0,
        "synthetic_clock_final": 11.0,
        "callback_trace": [
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "verify",
            "synthetic_time": 0.0,
            "effects": 0,
            "matched": false
          },
          {
            "callback": "candidates",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "scripted_selection",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 0
          },
          {
            "callback": "execute",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "observe",
            "synthetic_time": 0.0,
            "effects": 1
          },
          {
            "callback": "verify",
            "synthetic_time": 11.0,
            "effects": 1,
            "matched": false
          }
        ],
        "journal_events": [
          "start",
          "observed",
          "verified",
          "candidates",
          "decision",
          "freshness",
          "intent",
          "executed",
          "observed",
          "verified",
          "terminal"
        ],
        "journal_candidate_events": 1,
        "request_sizes": [
          {
            "selection_input_bytes": 1684,
            "shared_context_bytes": 272
          }
        ],
        "native_inference_attempts": 0,
        "real_network_attempts": 0
      },
      "correct": true,
      "whole_operation_seconds": 0.0017298329912591726,
      "returned_context_bytes": 651,
      "actual_model_usage": {
        "requests": 0,
        "models": {},
        "input_tokens": 0,
        "output_tokens": 0,
        "usage_complete": true
      },
      "arm": "candidate",
      "repeat": 3
    }
  ]
}