{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "title": "Edxperimental Labs Benchmark Run Intake Row",
  "type": "object",
  "additionalProperties": false,
  "required": [
    "suiteSlug",
    "taskId",
    "modelId",
    "provider",
    "runSeed",
    "startedAt",
    "promptPacketHash",
    "inputArtifactUri",
    "outputArtifactUri",
    "toolTraceUri",
    "screenshotUri",
    "rawCostUsd",
    "latencyMs",
    "score",
    "reviewer",
    "reviewerNote"
  ],
  "properties": {
    "suiteSlug": {
      "type": "string",
      "description": "Existing benchmark suite slug, for example indian-enterprise-workflow-suite."
    },
    "taskId": {
      "type": "string",
      "description": "Stable task id that matches the task packet or a new proposed packet."
    },
    "modelId": {
      "type": "string",
      "description": "Exact model or agent identifier, including provider version when available."
    },
    "provider": {
      "type": "string",
      "description": "Model/API/provider or local stack used for the run."
    },
    "runSeed": {
      "type": "string",
      "description": "Deterministic seed or run attempt label."
    },
    "startedAt": {
      "type": "string",
      "description": "ISO timestamp for the run start."
    },
    "promptPacketHash": {
      "type": "string",
      "description": "Hash or immutable id for the prompt/task packet shown to the model."
    },
    "inputArtifactUri": {
      "type": "string",
      "description": "Local path, object key, or redacted source packet id for the input."
    },
    "outputArtifactUri": {
      "type": "string",
      "description": "Local path, object key, or redacted output packet id for the model answer."
    },
    "toolTraceUri": {
      "type": "string",
      "description": "Tool-call trace, browser trace, terminal log, or not_applicable."
    },
    "screenshotUri": {
      "type": "string",
      "description": "Browser/app screenshot proof or not_applicable."
    },
    "rawCostUsd": {
      "type": "string",
      "description": "Observed provider cost for this run when available."
    },
    "latencyMs": {
      "type": "string",
      "description": "Wall-clock latency from run start to final answer."
    },
    "score": {
      "type": "string",
      "description": "Reviewer score on the suite rubric, 0 to 100."
    },
    "reviewer": {
      "type": "string",
      "description": "Human reviewer or review queue owner."
    },
    "reviewerNote": {
      "type": "string",
      "description": "Short note explaining acceptance, partial credit, or failure."
    }
  }
}
