{
  "schema_version": "0.13",
  "skill": {
    "name": "amc-run-video-calibration",
    "version": "unknown",
    "content_hash": null
  },
  "suite": {
    "format": "nvidia/benchmark-card",
    "suite_hash": null,
    "case_count": 5
  },
  "run": {
    "model_id": "openai/openai/gpt-5.5",
    "model_release_date": null,
    "provider": "openai",
    "surface": "external",
    "source": "imported/nvidia-cards",
    "runner_version": "0.21.0",
    "date_utc": "2026-08-08T00:00:00Z",
    "registry": "unregistered",
    "transcripts": "none",
    "judge": {
      "temperature": null,
      "sampling": "external",
      "surface": "external",
      "model_id": "unknown",
      "prompt_template_hash": null
    },
    "answered_by": {
      "kind": "external",
      "attested": false,
      "reported_model": null,
      "reported_models": null,
      "isolation": "none"
    },
    "harness": {
      "name": "codex",
      "version": null
    },
    "import": {
      "tool": "nvidia-cards",
      "format": "BENCHMARK.md",
      "format_version": null,
      "source_sha256": "d0507a7f1cf5265ff4b460faeebf0f6c38eabe074525bf436c6a301ae80a3a21",
      "imported_at": "2026-10-14T10:50:48.771Z",
      "comparison_method": "matched-measured-cases/1",
      "sidecars": [
        {
          "file": "benchmarks.json",
          "sha256": "e227f5baf9f654725a2ab697ed4b18d0fb8b58a03d2923d0d451f0dbc2dc258f"
        }
      ],
      "notices": [
        "figures are carried as printed in the card with sha256 d0507a7f1cf5265ff4b460faeebf0f6c38eabe074525bf436c6a301ae80a3a21, in source_figures; Driftproof records the files it read and does not attest the author; aggregates only: the card gives one score per dimension and agent and no per-run scores, so no band can be formed and no verdict is read from this receipt",
        "the source is recorded as NVIDIA/skills at commit 14a98ae278cec7e867695189d0f4003c08e4f765: the path and git blob sha1 of the card and of benchmarks.json are hashed into this receipt in source_figures.source, so each file can be checked against that commit",
        "run.model_id is the model the card names for codex",
        "the card gives the evaluation date to the day: run.date_utc is the start of that day, source_figures.evaluation_date is the date as printed, and readers show the day alone",
        "the card has no token table: economics are unknown",
        "the card's own verdict, tier status, findings and recommendation are not carried",
        "the card does not state skill.version, skill.content_hash, suite.suite_hash, run.harness.version, run.judge.model_id: each is unknown or null"
      ]
    }
  },
  "results": {
    "cases": [
      {
        "id": "dimension:security",
        "mode": "with_skill",
        "label": "Security",
        "case_status": "no_observations"
      },
      {
        "id": "dimension:security",
        "mode": "baseline",
        "label": "Security",
        "case_status": "no_observations"
      },
      {
        "id": "dimension:correctness",
        "mode": "with_skill",
        "label": "Correctness",
        "case_status": "no_observations"
      },
      {
        "id": "dimension:correctness",
        "mode": "baseline",
        "label": "Correctness",
        "case_status": "no_observations"
      },
      {
        "id": "dimension:discoverability",
        "mode": "with_skill",
        "label": "Discoverability",
        "case_status": "no_observations"
      },
      {
        "id": "dimension:discoverability",
        "mode": "baseline",
        "label": "Discoverability",
        "case_status": "no_observations"
      },
      {
        "id": "dimension:effectiveness",
        "mode": "with_skill",
        "label": "Effectiveness",
        "case_status": "no_observations"
      },
      {
        "id": "dimension:effectiveness",
        "mode": "baseline",
        "label": "Effectiveness",
        "case_status": "no_observations"
      },
      {
        "id": "dimension:efficiency",
        "mode": "with_skill",
        "label": "Efficiency",
        "case_status": "no_observations"
      },
      {
        "id": "dimension:efficiency",
        "mode": "baseline",
        "label": "Efficiency",
        "case_status": "no_observations"
      }
    ],
    "aggregates": {
      "with_skill": {
        "case_count": 0,
        "pass_count": 0,
        "borderline_count": 0,
        "mean_score": null,
        "stddev": null
      },
      "baseline": {
        "case_count": 0,
        "pass_count": 0,
        "borderline_count": 0,
        "mean_score": null,
        "stddev": null
      },
      "band_rule": "per arm: the sample standard deviation (n-1) of the per-case means across the included cases; the comparison band is the quadrature sum of the two arms; null where an arm has fewer than two included cases",
      "excluded_cases": [
        {
          "id": "dimension:security",
          "modes": [
            "with_skill",
            "baseline"
          ],
          "reason": "aggregates only: the source carries no per-run scores for this case, so no band can be formed"
        },
        {
          "id": "dimension:correctness",
          "modes": [
            "with_skill",
            "baseline"
          ],
          "reason": "aggregates only: the source carries no per-run scores for this case, so no band can be formed"
        },
        {
          "id": "dimension:discoverability",
          "modes": [
            "with_skill",
            "baseline"
          ],
          "reason": "aggregates only: the source carries no per-run scores for this case, so no band can be formed"
        },
        {
          "id": "dimension:effectiveness",
          "modes": [
            "with_skill",
            "baseline"
          ],
          "reason": "aggregates only: the source carries no per-run scores for this case, so no band can be formed"
        },
        {
          "id": "dimension:efficiency",
          "modes": [
            "with_skill",
            "baseline"
          ],
          "reason": "aggregates only: the source carries no per-run scores for this case, so no band can be formed"
        }
      ]
    }
  },
  "comparison": {
    "with_skill_score": null,
    "baseline_score": null,
    "delta": null,
    "delta_uncertainty": null,
    "delta_uncertainty_unavailable": "no_cases"
  },
  "verification_level": "DECLARED",
  "uncertainty_unavailable": "aggregates_only",
  "source_figures": {
    "agent": "codex",
    "skill": "amc-run-video-calibration",
    "evaluation_date": "2026-08-08",
    "evaluator_version": "1.1.2",
    "environment": "k8s-sandbox",
    "dataset_digest": "sha256:27e110482838494c58572fb5c4852aa0ba9c36f0b5f423ee26f53e1a25ed3ed0",
    "tasks": 6,
    "attempts_per_task": 1,
    "source": {
      "repo": "NVIDIA/skills",
      "commit": "14a98ae278cec7e867695189d0f4003c08e4f765",
      "files": [
        {
          "path": "skills/amc-run-video-calibration/BENCHMARK.md",
          "git_blob_sha1": "39b753c017d808f7302942917c59cb08ddc389c7"
        },
        {
          "path": "benchmarks.json",
          "git_blob_sha1": "da94ea53fb6868b939cd49b1b14f9f97f78e2976"
        }
      ]
    },
    "rows": [
      {
        "dimension": "Overall",
        "printed": "35% → 83% (+48 points)",
        "score_pct": 83,
        "baseline_pct": 35,
        "uplift": 48,
        "uplift_unit": "points",
        "tasks": null
      },
      {
        "dimension": "Security",
        "printed": "42% → 100% (+58 points)",
        "score_pct": 100,
        "baseline_pct": 42,
        "uplift": 58,
        "uplift_unit": "points",
        "tasks": null
      },
      {
        "dimension": "Correctness",
        "printed": "13% → 77% (+63 points)",
        "score_pct": 77,
        "baseline_pct": 13,
        "uplift": 63,
        "uplift_unit": "points",
        "tasks": null
      },
      {
        "dimension": "Discoverability",
        "printed": "47% → 89% (+42 points)",
        "score_pct": 89,
        "baseline_pct": 47,
        "uplift": 42,
        "uplift_unit": "points",
        "tasks": null
      },
      {
        "dimension": "Effectiveness",
        "printed": "27% → 49% (+21 points)",
        "score_pct": 49,
        "baseline_pct": 27,
        "uplift": 21,
        "uplift_unit": "points",
        "tasks": null
      },
      {
        "dimension": "Efficiency",
        "printed": "47% → 100% (+53 points)",
        "score_pct": 100,
        "baseline_pct": 47,
        "uplift": 53,
        "uplift_unit": "points",
        "tasks": null
      }
    ],
    "tokens": null
  },
  "receipt_hash": "55a864ccf4785f862e6615d64ee58d98c6df576e3b56dd40cb5783da52be59ec"
}