{
  "captured_from": "https://marufov.com/api/lab/run/wrun_01M3WXWPKMF3H8Q66KZA56MZCV",
  "captured_at": "2026-10-02",
  "note": "The status route's response for one public run on production, stored byte-for-byte below so the homepage can show a real run without starting one. The run was started from /lab with the default preset, count-guard-v1.",
  "preset": "count-guard-v1",
  "response": {
    "run_id": "wrun_01M3WXWPKMF3H8Q66KZA56MZCV",
    "status": "completed",
    "finished": true,
    "outcome": {
      "kind": "graded",
      "verdict": "rejected",
      "reason": "association-exact: 0/3 cases satisfied, threshold 1",
      "grading": {
        "verdict": "rejected",
        "reason": "association-exact: 0/3 cases satisfied, threshold 1",
        "spec_version": 2,
        "spec_hash": "f027762ab4d08b35",
        "declared_protocol": "positional-v0",
        "criteria": [
          {
            "id": "universal-refusal",
            "question": "Does it refuse every response from which no association can be recovered?",
            "applicable": 48,
            "satisfied": 48,
            "passed": true
          },
          {
            "id": "association-exact",
            "question": "On its own protocol, does every article receive exactly the verdict it was given?",
            "applicable": 3,
            "satisfied": 0,
            "passed": false
          },
          {
            "id": "protocol-violation-refusal",
            "question": "Does it refuse duplicate, unknown, missing ids and unusable scores?",
            "applicable": 1,
            "satisfied": 0,
            "passed": false
          },
          {
            "id": "no-crash",
            "question": "Does it terminate on every case without crashing or hanging?",
            "applicable": 52,
            "satisfied": 52,
            "passed": true
          },
          {
            "id": "complete-evidence",
            "question": "Is there a prediction record for every applicable case?",
            "applicable": 52,
            "satisfied": 52,
            "passed": true
          },
          {
            "id": "protocol-exclusivity",
            "question": "On cases outside its declared protocol, does it refuse rather than associate anyway?",
            "applicable": 12,
            "satisfied": 5,
            "passed": false
          }
        ],
        "counts": {
          "correct": 48,
          "wrong-association": 3,
          "should-have-refused": 1,
          "should-have-parsed": 0,
          "crashed": 0,
          "timeout": 0,
          "missing-record": 0,
          "not-applicable": 12
        },
        "out_of_protocol_case_ids": [
          "syn-keyed-in-order",
          "syn-keyed-reversed",
          "syn-keyed-rotated",
          "syn-unknown-id",
          "syn-score-nan",
          "syn-score-out-of-range",
          "syn-relevant-not-bool"
        ],
        "cases": [
          {
            "case_id": "observed-2026-09-02-000",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "27 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-001",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-002",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "18 verdicts recorded for 20 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-003",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-004",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-005",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "protocol-association",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "wrong-association",
            "detail": "no ground truth recorded",
            "why": "entry count matches the batch; association is parseable but unverified",
            "expected_refusal_kinds": [],
            "observed_refusal_kind": null
          },
          {
            "case_id": "observed-2026-09-02-006",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "201 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-007",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "33 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-008",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "37 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-009",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-010",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "19 verdicts recorded for 20 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-011",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-012",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-013",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "25 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-014",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-015",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "19 verdicts recorded for 20 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-016",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-017",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-018",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "59 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-019",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-020",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "19 verdicts recorded for 20 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-021",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-022",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-023",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "36 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-024",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "21 verdicts recorded for 20 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-025",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "31 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-026",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-027",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "19 verdicts recorded for 20 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-028",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "38 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-029",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-030",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-031",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "31 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-032",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "53 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-033",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "19 verdicts recorded for 20 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-034",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "34 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-035",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "protocol-association",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "wrong-association",
            "detail": "no ground truth recorded",
            "why": "entry count matches the batch; association is parseable but unverified",
            "expected_refusal_kinds": [],
            "observed_refusal_kind": null
          },
          {
            "case_id": "observed-2026-09-02-036",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-037",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-038",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "completion stopped for 'length'",
            "expected_refusal_kinds": [
              "truncated_response",
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "observed-2026-09-02-039",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "25 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-040",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "254 verdicts recorded for 40 articles",
            "expected_refusal_kinds": [
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "observed-2026-09-02-041",
            "group": "observed",
            "origin": "recorded-replay",
            "family": "protocol-association",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "wrong-association",
            "detail": "no ground truth recorded",
            "why": "entry count matches the batch; association is parseable but unverified",
            "expected_refusal_kinds": [],
            "observed_refusal_kind": null
          },
          {
            "case_id": "syn-keyed-in-order",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "protocol-association",
            "protocol": "keyed-v2",
            "applicability": "not-applicable",
            "status": "not-applicable",
            "detail": "case is keyed-v2; candidate declares positional-v0",
            "why": "a well-formed keyed response in request order",
            "expected_refusal_kinds": [],
            "observed_refusal_kind": null
          },
          {
            "case_id": "syn-keyed-reversed",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "protocol-association",
            "protocol": "keyed-v2",
            "applicability": "not-applicable",
            "status": "not-applicable",
            "detail": "case is keyed-v2; candidate declares positional-v0",
            "why": "the same verdicts, reversed; association must not move",
            "expected_refusal_kinds": [],
            "observed_refusal_kind": null
          },
          {
            "case_id": "syn-keyed-rotated",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "protocol-association",
            "protocol": "keyed-v2",
            "applicability": "not-applicable",
            "status": "not-applicable",
            "detail": "case is keyed-v2; candidate declares positional-v0",
            "why": "the same verdicts, rotated; association must not move",
            "expected_refusal_kinds": [],
            "observed_refusal_kind": null
          },
          {
            "case_id": "syn-duplicate-id",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "protocol-association",
            "protocol": "keyed-v2",
            "applicability": "not-applicable",
            "status": "not-applicable",
            "detail": "case is keyed-v2; candidate declares positional-v0",
            "why": "the same article is judged twice; no winner may be picked",
            "expected_refusal_kinds": [
              "duplicate_id"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "syn-unknown-id",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "protocol-association",
            "protocol": "keyed-v2",
            "applicability": "not-applicable",
            "status": "not-applicable",
            "detail": "case is keyed-v2; candidate declares positional-v0",
            "why": "a verdict for an article that was never sent",
            "expected_refusal_kinds": [
              "unknown_id"
            ],
            "observed_refusal_kind": null
          },
          {
            "case_id": "syn-missing-id",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "protocol-association",
            "protocol": "keyed-v2",
            "applicability": "not-applicable",
            "status": "not-applicable",
            "detail": "case is keyed-v2; candidate declares positional-v0",
            "why": "one article receives no verdict; partial salvage is not allowed",
            "expected_refusal_kinds": [
              "missing_id",
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "syn-score-nan",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "protocol-association",
            "protocol": "keyed-v2",
            "applicability": "not-applicable",
            "status": "not-applicable",
            "detail": "case is keyed-v2; candidate declares positional-v0",
            "why": "score is nan; production clamps it into a real verdict",
            "expected_refusal_kinds": [
              "score_not_finite",
              "invalid_type",
              "malformed_json"
            ],
            "observed_refusal_kind": null
          },
          {
            "case_id": "syn-score-out-of-range",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "protocol-association",
            "protocol": "keyed-v2",
            "applicability": "not-applicable",
            "status": "not-applicable",
            "detail": "case is keyed-v2; candidate declares positional-v0",
            "why": "score is out-of-range; production clamps it into a real verdict",
            "expected_refusal_kinds": [
              "score_out_of_range",
              "invalid_type"
            ],
            "observed_refusal_kind": null
          },
          {
            "case_id": "syn-score-string",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "protocol-association",
            "protocol": "keyed-v2",
            "applicability": "not-applicable",
            "status": "not-applicable",
            "detail": "case is keyed-v2; candidate declares positional-v0",
            "why": "score is string; production clamps it into a real verdict",
            "expected_refusal_kinds": [
              "invalid_type"
            ],
            "observed_refusal_kind": "invalid_type"
          },
          {
            "case_id": "syn-score-null",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "protocol-association",
            "protocol": "keyed-v2",
            "applicability": "not-applicable",
            "status": "not-applicable",
            "detail": "case is keyed-v2; candidate declares positional-v0",
            "why": "score is null; production clamps it into a real verdict",
            "expected_refusal_kinds": [
              "invalid_type"
            ],
            "observed_refusal_kind": "invalid_type"
          },
          {
            "case_id": "syn-relevant-not-bool",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "protocol-association",
            "protocol": "keyed-v2",
            "applicability": "not-applicable",
            "status": "not-applicable",
            "detail": "case is keyed-v2; candidate declares positional-v0",
            "why": "relevant is a string",
            "expected_refusal_kinds": [
              "invalid_type"
            ],
            "observed_refusal_kind": null
          },
          {
            "case_id": "syn-malformed-json",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "universal-refusal",
            "protocol": "keyed-v2",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (malformed_json)",
            "why": "the response is cut mid-object",
            "expected_refusal_kinds": [
              "malformed_json"
            ],
            "observed_refusal_kind": "malformed_json"
          },
          {
            "case_id": "syn-truncated-finish-reason",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "universal-refusal",
            "protocol": "keyed-v2",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "the completion stopped at the output ceiling",
            "expected_refusal_kinds": [
              "truncated_response",
              "count_mismatch",
              "missing_id"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "syn-not-an-object",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "universal-refusal",
            "protocol": "keyed-v2",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (unexpected_shape)",
            "why": "top level is an array, not an object",
            "expected_refusal_kinds": [
              "unexpected_shape"
            ],
            "observed_refusal_kind": "unexpected_shape"
          },
          {
            "case_id": "syn-empty-results",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "protocol-association",
            "protocol": "keyed-v2",
            "applicability": "not-applicable",
            "status": "not-applicable",
            "detail": "case is keyed-v2; candidate declares positional-v0",
            "why": "no verdicts at all",
            "expected_refusal_kinds": [
              "missing_id",
              "count_mismatch"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "syn-exec-no-recording",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "universal-refusal",
            "protocol": "keyed-v2",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (no_recording)",
            "why": "execution ended as no_recording; an unavailable answer is not an answer",
            "expected_refusal_kinds": [
              "no_recording"
            ],
            "observed_refusal_kind": "no_recording"
          },
          {
            "case_id": "syn-exec-timeout",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "universal-refusal",
            "protocol": "keyed-v2",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (timeout)",
            "why": "execution ended as timeout; an unavailable answer is not an answer",
            "expected_refusal_kinds": [
              "timeout"
            ],
            "observed_refusal_kind": "timeout"
          },
          {
            "case_id": "syn-exec-cancelled",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "universal-refusal",
            "protocol": "keyed-v2",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (cancelled)",
            "why": "execution ended as cancelled; an unavailable answer is not an answer",
            "expected_refusal_kinds": [
              "cancelled"
            ],
            "observed_refusal_kind": "cancelled"
          },
          {
            "case_id": "syn-exec-budget",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "universal-refusal",
            "protocol": "keyed-v2",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (retries_exhausted)",
            "why": "execution ended as budget_exceeded; an unavailable answer is not an answer",
            "expected_refusal_kinds": [
              "retries_exhausted",
              "no_recording"
            ],
            "observed_refusal_kind": "retries_exhausted"
          },
          {
            "case_id": "syn-positional-short",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "4 verdicts for 6 articles: which belongs to which is unrecorded",
            "expected_refusal_kinds": [
              "count_mismatch",
              "missing_id"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "syn-positional-long",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "universal-refusal",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "correct",
            "detail": "refused (count_mismatch)",
            "why": "7 verdicts for 6 articles",
            "expected_refusal_kinds": [
              "count_mismatch",
              "duplicate_id"
            ],
            "observed_refusal_kind": "count_mismatch"
          },
          {
            "case_id": "syn-positional-reordered",
            "group": "synthetic",
            "origin": "fault-injection",
            "family": "protocol-association",
            "protocol": "positional-v0",
            "applicability": "scored",
            "status": "should-have-refused",
            "detail": "equal length, internally reordered — a length check cannot detect this",
            "why": "equal length, internally reordered — a length check cannot detect this",
            "expected_refusal_kinds": [
              "count_mismatch",
              "unexpected_shape",
              "unsupported_protocol",
              "invalid_type",
              "missing_id"
            ],
            "observed_refusal_kind": null
          }
        ]
      },
      "sandbox": {
        "sandbox_id": "azure-sympathetic-flyingfish-Wap4Q6",
        "region": "iad1",
        "runtime": "python3.13",
        "network_policy": "deny-all",
        "boot_ms": 230,
        "wall_clock_ms": 546,
        "active_cpu_ms": 3223,
        "egress_bytes": 34949,
        "exit_code": 0,
        "isolation": [
          {
            "name": "egress-dns",
            "expectation": "DNS resolution of an external host fails",
            "held": true
          },
          {
            "name": "egress-https",
            "expectation": "an outbound HTTPS request fails",
            "held": true
          },
          {
            "name": "no-evaluator-present",
            "expectation": "the code that grades the candidate is nowhere on the filesystem",
            "held": true
          },
          {
            "name": "no-secrets-in-env",
            "expectation": "no production credential is present in the environment",
            "held": true
          }
        ]
      },
      "events": [
        {
          "type": "created",
          "run_id": "visitor__muq6q1w3",
          "candidate_id": "visitor",
          "at": "2026-10-01T23:48:16.578Z",
          "spec_hash": "f027762ab4d08b35"
        },
        {
          "type": "scope-checked",
          "at": "2026-10-01T23:48:16.578Z",
          "allowed": true,
          "detail": "backend/lab/contract/candidate.py is within the allowed patch scope"
        },
        {
          "type": "attempt-started",
          "attempt_id": "visitor__muq6q1w3#01",
          "runner": "vercel-sandbox",
          "at": "2026-10-01T23:48:16.613Z"
        },
        {
          "type": "attempt-ended",
          "attempt_id": "visitor__muq6q1w3#01",
          "at": "2026-10-01T23:48:23.338Z",
          "status": "succeeded",
          "note": "sandbox azure-sympathetic-flyingfish-Wap4Q6 in iad1, 34949 bytes egress"
        },
        {
          "type": "evaluated",
          "at": "2026-10-01T23:48:24.155Z",
          "verdict": "rejected",
          "reason": "association-exact: 0/3 cases satisfied, threshold 1"
        }
      ],
      "processes": [
        {
          "step": "scope",
          "process_id": "dpl_6A5Z7rXB16iYQb8vk4BsAupvssDe:68bbaf1e",
          "at": "2026-10-01T23:48:16.578Z",
          "uptime_s": 15
        },
        {
          "step": "execute",
          "process_id": "dpl_6A5Z7rXB16iYQb8vk4BsAupvssDe:68bbaf1e",
          "at": "2026-10-01T23:48:23.338Z",
          "uptime_s": 21.8
        },
        {
          "step": "grade",
          "process_id": "dpl_6A5Z7rXB16iYQb8vk4BsAupvssDe:68bbaf1e",
          "at": "2026-10-01T23:48:24.155Z",
          "uptime_s": 22.6
        }
      ],
      "resumed": false
    },
    "progress": [
      {
        "at": "2026-10-01T23:48:16.571Z",
        "stage": "scope checked: only candidate.py is written"
      },
      {
        "at": "2026-10-01T23:48:17.410Z",
        "stage": "creating microVM"
      },
      {
        "at": "2026-10-01T23:48:17.640Z",
        "stage": "uploading 5 files"
      },
      {
        "at": "2026-10-01T23:48:17.767Z",
        "stage": "running the harness"
      },
      {
        "at": "2026-10-01T23:48:18.313Z",
        "stage": "probing isolation"
      },
      {
        "at": "2026-10-01T23:48:19.177Z",
        "stage": "stopping the microVM"
      },
      {
        "at": "2026-10-01T23:48:23.337Z",
        "stage": "microVM stopped"
      }
    ]
  }
}