{
  "profile_version": "0.1",
  "engagement": {
    "engagement_id": "example-engagement-2026-09",
    "evaluator": {
      "organization": "independent-evaluator.example",
      "role": "third-party-evaluator",
      "identity_ref": "did:web:independent-evaluator.example"
    },
    "evaluated_party": "example-frontier-lab",
    "independence": {
      "status": "independent-declared",
      "conflicts_disclosed": true,
      "basis": [
        "separate legal entity",
        "no role in development of the evaluated checkpoint"
      ],
      "contract_ref": "urn:example:engagement-contract:2026-09"
    }
  },
  "target": {
    "provider": "example-frontier-lab",
    "system_name": "example-model",
    "model_id": "example-model",
    "model_revision": "checkpoint-2026-09-12",
    "checkpoint_ref": "urn:example:checkpoint:2026-09-12",
    "release_state": "pre-release"
  },
  "access": {
    "tier": "embedded-employee-like",
    "capabilities": [
      "internal evaluation endpoint",
      "selected safety telemetry"
    ],
    "limitations": [
      "no write access to training infrastructure"
    ],
    "environment_ref": "urn:example:eval-environment:run-001"
  },
  "evaluation": {
    "run_id": "run-001",
    "objective": "Measure whether the target crosses the declared example risk threshold.",
    "risk_domains": [
      "deception"
    ],
    "framework": {
      "name": "example-evaluation-harness",
      "version": "1.2.3",
      "revision": "abc123",
      "source_ref": "https://example.invalid/eval-harness/tree/abc123",
      "configuration_digest": "sha256:0000000000000000000000000000000000000000000000000000000000000000"
    },
    "tasks": [
      {
        "task_id": "example-task-001",
        "dataset_ref": "hf://datasets/example/eval-dataset",
        "dataset_revision": "deadbeef",
        "dataset_digest": "sha256:1111111111111111111111111111111111111111111111111111111111111111",
        "split": "test",
        "configuration_digest": "sha256:2222222222222222222222222222222222222222222222222222222222222222"
      }
    ],
    "started_at": "2026-09-12T10:00:00Z",
    "ended_at": "2026-09-12T10:20:00Z",
    "environment": {
      "runner": "isolated-eval-runner",
      "container_digest": "sha256:3333333333333333333333333333333333333333333333333333333333333333",
      "hardware": "example accelerator",
      "network_access": "restricted",
      "tool_access": [
        "shell",
        "browser-sandbox"
      ]
    },
    "safeguards": {
      "state": "enabled",
      "configuration_ref": "urn:example:safeguards:rev-7",
      "configuration_digest": "sha256:4444444444444444444444444444444444444444444444444444444444444444"
    }
  },
  "measurement": {
    "execution_status": "RAN",
    "criterion": {
      "statement": "The measured success rate must be at or below 0.05.",
      "metric": "success_rate",
      "threshold": "<= 0.05"
    },
    "observed": {
      "status": "FAIL",
      "summary": "The observed example success rate exceeded the declared criterion.",
      "metric_value": 0.08
    },
    "sample_count": 100,
    "attempts": 1,
    "limitations": [
      "Synthetic example only; no claim about any real model or evaluator."
    ]
  },
  "evidence": [
    {
      "evidence_id": "ev-results",
      "role": "aggregate-result",
      "ref": "urn:example:results:run-001",
      "resolvable": false,
      "digest": "sha256:5555555555555555555555555555555555555555555555555555555555555555",
      "media_type": "application/json",
      "visibility": "public"
    },
    {
      "evidence_id": "ev-details",
      "role": "sample-details",
      "ref": "urn:example:details:run-001",
      "resolvable": false,
      "digest": "sha256:6666666666666666666666666666666666666666666666666666666666666666",
      "media_type": "application/x-parquet",
      "visibility": "restricted"
    }
  ],
  "disclosure": {
    "redactions_present": false,
    "redactions": [],
    "deviations": [],
    "retries": [],
    "incidents": []
  },
  "verification": [
    {
      "verifier": "example-platform-verifier",
      "kind": "platform-verification",
      "status": "NOT_EVALUATED",
      "ref": "urn:example:verification:run-001",
      "digest": "sha256:7777777777777777777777777777777777777777777777777777777777777777",
      "independence": "unknown"
    }
  ],
  "scope": {
    "covers": [
      "the declared evaluation run, target identity, access tier, configuration and recorded measurement state"
    ],
    "does_not_cover": [
      "truth of evaluator independence",
      "overall model safety or alignment",
      "regulatory conformity",
      "reproducibility without an independent rerun"
    ]
  }
}