TradeOS Dataset Concierge

{
  "input": {
    "body": {
      "output": "json",
      "symbol": "BTC",
      "chain_id": "ethereum",
      "metadata": {
        "use_case": "risk-evaluation",
        "workflow_type": "calibration"
      },
      "contract_address": "0x0000000000000000000000000000000000000000"
    },
    "type": "http",
    "method": "POST",
    "bodyType": "json"
  }
}

{
  "details": {
    "packages": [
      {
        "package_id": "tradeos-decision-trajectory",
        "access_mode": "private_entitlement_required",
        "description": "Point-in-time decision trajectories across signal, modifier, fusion, feasibility, and outcome context.",
        "builder_value": "Train or evaluate agents that need to explain why a setup passed, paused for review, or failed controls."
      },
      {
        "package_id": "tradeos-forecast-calibration",
        "access_mode": "private_entitlement_required",
        "description": "Forecast envelopes, horizon outcomes, hit/progress labels, and calibration slices.",
        "builder_value": "Benchmark forecast-aware agents without treating forecasts as trading instructions."
      },
      {
        "package_id": "tradeos-replay-audit",
        "access_mode": "private_entitlement_required",
        "description": "Replay findings, model drift notes, weak-horizon cohorts, and regression evidence.",
        "builder_value": "Build evaluation harnesses and regression tests for trading-intelligence agents."
      },
      {
        "package_id": "tradeos-risk-intervention",
        "access_mode": "private_entitlement_required",
        "description": "Drawdown, breaker, risk-event, and safety-control context around decision windows.",
        "builder_value": "Evaluate whether builder agents respect risk constraints before surfacing downstream actions."
      }
    ],
    "access_path": {
      "next_step": "Send the requested package, intended use case, and entitlement needs for access review.",
      "support_contact": "tradeos.contact@gmail.com"
    },
    "human_summary": {
      "headline": "Curated TradeOS dataset packages are available for access review.",
      "data_gaps": [],
      "builder_use": "Use to select the right dataset package and start access review; do not treat the public agent as a raw dataset download path.",
      "score_notes": [
        "Public boundary: catalog, package fit, schema guidance, and access path. Private boundary: raw data delivery and enterprise terms."
      ],
      "package_cards": [
        {
          "package": "tradeos-decision-trajectory",
          "best_for": "Train or evaluate agents that need to explain why a setup passed, paused for review, or failed controls.",
          "includes": "Point-in-time decision trajectories across signal, modifier, fusion, feasibility, and outcome context.",
          "next_step": "Email tradeos.contact@gmail.com with package, intended use case, schema/sample needs, and entitlement requirements.",
          "access_mode": "Private entitlement required."
        },
        {
          "package": "tradeos-forecast-calibration",
          "best_for": "Benchmark forecast-aware agents without treating forecasts as trading instructions.",
          "includes": "Forecast envelopes, horizon outcomes, hit/progress labels, and calibration slices.",
          "next_step": "Email tradeos.contact@gmail.com with package, intended use case, schema/sample needs, and entitlement requirements.",
          "access_mode": "Private entitlement required."
        },
        {
          "package": "tradeos-replay-audit",
          "best_for": "Build evaluation harnesses and regression tests for trading-intelligence agents.",
          "includes": "Replay findings, model drift notes, weak-horizon cohorts, and regression evidence.",
          "next_step": "Email tradeos.contact@gmail.com with package, intended use case, schema/sample needs, and entitlement requirements.",
          "access_mode": "Private entitlement required."
        },
        {
          "package": "tradeos-risk-intervention",
          "best_for": "Evaluate whether builder agents respect risk constraints before surfacing downstream actions.",
          "includes": "Drawdown, breaker, risk-event, and safety-control context around decision windows.",
          "next_step": "Email tradeos.contact@gmail.com with package, intended use case, schema/sample needs, and entitlement requirements.",
          "access_mode": "Private entitlement required."
        }
      ],
      "verdict_label": "Curated dataset catalog available",
      "support_contact": "tradeos.contact@gmail.com",
      "product_primitive": "Dataset concierge",
      "what_needs_review": [
        "Raw private data delivery requires entitlement, redistribution, and data-rights review."
      ],
      "what_is_constructive": [
        "4 dataset package cards are available.",
        "Agent evaluation and benchmark harnesses.",
        "Forecast calibration and horizon-quality research."
      ],
      "selected_route_reason": "Selected because the request resolved to dataset concierge: package fit and access-path guidance.",
      "plain_language_summary": "The public agent describes package fit, schemas, boundaries, and access path. Raw/private datasets are not self-serve through AntSeed."
    },
    "delivery_boundary": {
      "public_agent": "catalog, schema guidance, package fit, access path",
      "private_contract": "raw or private dataset delivery, redistribution terms, enterprise SLAs"
    },
    "best_fit_use_cases": [
      "agent evaluation and benchmark harnesses",
      "forecast calibration and horizon-quality research",
      "signal-decision replay and human-review workflows",
      "risk-intervention and safety-control analysis"
    ],
    "reppo_pipeline_configured": false
  },
  "drivers": [
    {
      "detail": "Catalog covers derived TradeOS datasets for builders and AI/quant evaluation.",
      "impact": "positive"
    },
    {
      "detail": "Raw private data access requires a separate entitlement and data-rights review.",
      "impact": "neutral"
    }
  ],
  "notices": [
    "Research output only. Not investment advice. No execution. No custody.",
    "Dataset concierge is an access and schema guidance surface; it does not distribute raw private data."
  ],
  "service": "tradeos-crypto-intelligence-agent",
  "subject": {
    "limit": null,
    "symbol": "BTC",
    "horizon": null,
    "chain_id": "ethereum",
    "metadata": {
      "use_case": "risk-evaluation",
      "marketplace": "generic_x402",
      "workflow_type": "calibration",
      "x402_provider": "x402",
      "x402_channel_id": "x402-direct",
      "requested_service": "tradeos-dataset-concierge",
      "x402_resource_type": "intelligence_report"
    },
    "timeframe": null,
    "contract_address": "0x0000000000000000000000000000000000000000"
  },
  "verdict": "curated_dataset_catalog_available",
  "evidence": [
    {
      "as_of": null,
      "detail": "4 curated dataset package(s) are described for builder discovery.",
      "impact": "neutral",
      "source": "tradeos-dataset-catalog",
      "severity": null,
      "reference": null
    }
  ],
  "confidence": 0.8,
  "created_at": "2026-06-18T05:12:31.409997Z",
  "request_id": "5116ca44-b7e0-4fb3-a0fe-323e8a73f0c7",
  "risk_flags": [
    "no_raw_dataset_download",
    "entitlement_required_for_private_data"
  ],
  "next_checks": [
    "Choose the dataset package and intended use case before requesting access.",
    "Confirm whether the buyer needs schemas, sample rows, evaluation reports, or a private delivery agreement.",
    "Keep raw private dataset download outside the public AntSeed agent until entitlements are approved."
  ],
  "missing_data": [],
  "request_type": "dataset_concierge",
  "response_mode": "json",
  "data_freshness": {
    "as_of": null,
    "source": "tradeos-dataset-catalog",
    "by_source": {},
    "freshness_known": false
  },
  "negative_drivers": [
    "Dataset delivery is not self-serve through this public agent.",
    "Redistribution rights and private data entitlements must be reviewed before transfer."
  ],
  "positive_drivers": [
    "Derived packages can support agent benchmarking, replay audit, calibration, and risk evaluation.",
    "Public AntSeed response provides schema/use-case guidance without exposing raw private datasets."
  ],
  "refusal_category": null
}