{
  "$schema": "https://themachinepress.com/schemas/story-v1.schema.json",
  "schema_version": "1.0.0",
  "document_type": "machine_press_story",
  "story": {
    "story_id": "mp-2026-09-10-007",
    "source_story_id": "tmp-story-contracteval-procedural-conformance",
    "edition_id": "mp-2026-09-10-morning-0063",
    "edition_url": "https://themachinepress.com/edition/2026-09-10",
    "position": 7,
    "story_type": "dispatch",
    "section": "safety-security",
    "editorial_classification": "editorial",
    "headline": "A Good Answer Could Still Skip the Required Check",
    "slug": "a-good-answer-could-still-skip-the-required-check",
    "dek": "ContractEval maps query-active procedural obligations to trace evidence and separates omissions, wrong branches and invariant breaches.",
    "summary": "ContractEval maps query-active procedural obligations to trace evidence and separates omissions, wrong branches and invariant breaches.",
    "body_text": "Output-only scoring can approve an answer even when the agent skipped the branch, check or dependency that justified it. ContractEval turns a procedure into obligations activated by the current query, then matches those obligations against a response or execution trace. Under gold expected and observed graphs, it localized every injected structural failure in a controlled audit suite; language-model extraction retained much of the signal but remained sensitive to calibration. The authors explicitly frame the method as an audit aid, not a compliance guarantee.",
    "why_it_matters": "ContractEval maps query-active procedural obligations to trace evidence and separates omissions, wrong branches and invariant breaches.",
    "limitations": [
      "Output-only scoring can approve an answer even when the agent skipped the branch, check or dependency that justified it.",
      "The authors explicitly frame the method as an audit aid, not a compliance guarantee."
    ],
    "importance": 8,
    "canonical_url": "https://themachinepress.com/story/mp-2026-09-10-007/a-good-answer-could-still-skip-the-required-check",
    "json_url": "https://themachinepress.com/story/mp-2026-09-10-007.json",
    "first_published_at": "2026-09-10T09:00:00.000-04:00",
    "modified_at": "2026-09-10T09:00:00.000-04:00",
    "content_status": "new",
    "is_carryover": false,
    "carryover_reason": null,
    "key_claims": [
      {
        "claim_id": "claim-mp-2026-09-10-007-001",
        "text": "ContractEval maps query-active procedural obligations to trace evidence and separates omissions, wrong branches and invariant breaches.",
        "source_ids": [
          "source-2026-09-10-007"
        ],
        "qualification": "Output-only scoring can approve an answer even when the agent skipped the branch, check or dependency that justified it."
      }
    ],
    "source_ids": [
      "source-2026-09-10-007"
    ],
    "tags": [
      "procedural compliance",
      "agent traces",
      "auditing"
    ],
    "image_url": null,
    "corrections": []
  },
  "sources": [
    {
      "source_id": "source-2026-09-10-007",
      "title": "arXiv preprint 2609.09458",
      "publisher": "arXiv",
      "url": "https://arxiv.org/abs/2609.09458",
      "canonical_url": "https://arxiv.org/abs/2609.09458",
      "source_type": "primary_research",
      "is_primary_source": true,
      "published_at": "2026-09-07T20:00:00.000-04:00",
      "accessed_at": "2026-09-10T08:24:00.000-04:00",
      "supports_claim_ids": [
        "claim-mp-2026-09-10-007-001"
      ]
    }
  ],
  "corrections": [],
  "publisher": {
    "name": "The Machine Press",
    "url": "https://themachinepress.com",
    "description": "A daily newspaper for the age of artificial intelligence."
  },
  "cite_this_report": {
    "title": "A Good Answer Could Still Skip the Required Check",
    "publisher": "The Machine Press",
    "published_at": "2026-09-10T09:00:00.000-04:00",
    "canonical_url": "https://themachinepress.com/story/mp-2026-09-10-007/a-good-answer-could-still-skip-the-required-check"
  }
}