{
  "$schema": "https://themachinepress.com/schemas/story-v1.schema.json",
  "schema_version": "1.0.0",
  "document_type": "machine_press_story",
  "story": {
    "story_id": "mp-2026-10-28-005",
    "source_story_id": "tmp-story-agent-handoff-continuity",
    "edition_id": "mp-2026-10-28-morning-0110",
    "edition_url": "https://themachinepress.com/edition/2026-10-28",
    "position": 5,
    "story_type": "dispatch",
    "section": "benchmarks-evals",
    "editorial_classification": "editorial",
    "headline": "A Handoff Has to Preserve Commitments, Not Just Facts",
    "slug": "a-handoff-has-to-preserve-commitments-not-just-facts",
    "dek": "Across 1,039 tasks, structured inherited state improved repository-evidence recovery for every tested receiver and continuation for most.",
    "summary": "Across 1,039 tasks, structured inherited state improved repository-evidence recovery for every tested receiver and continuation for most.",
    "body_text": "Long-running work often reaches a new model after earlier decisions, tool calls and file changes have already constrained what remains. HSC-Bench measures two distinct parts of that takeover across 1,039 tasks: recovering relevant repository evidence and continuing from established conversational, tool and file state. Six receiving models performed differently across the two tracks, showing that memory retrieval alone does not predict whether a model will honor completed work and prior commitments. Structured handoff state improved evidence recovery for every receiver and continuation performance for most. The benchmark does not prove a universal handoff format, but it makes continuity observable as downstream behavior rather than a summary-quality score.",
    "why_it_matters": "Across 1,039 tasks, structured inherited state improved repository-evidence recovery for every tested receiver and continuation for most.",
    "limitations": [
      "Six receiving models performed differently across the two tracks, showing that memory retrieval alone does not predict whether a model will honor completed work and prior commitments.",
      "The benchmark does not prove a universal handoff format, but it makes continuity observable as downstream behavior rather than a summary-quality score."
    ],
    "importance": 9,
    "canonical_url": "https://themachinepress.com/story/mp-2026-10-28-005/a-handoff-has-to-preserve-commitments-not-just-facts",
    "json_url": "https://themachinepress.com/story/mp-2026-10-28-005.json",
    "first_published_at": "2026-10-28T09:00:00.000-04:00",
    "modified_at": "2026-10-28T09:00:00.000-04:00",
    "content_status": "new",
    "is_carryover": false,
    "carryover_reason": null,
    "key_claims": [
      {
        "claim_id": "claim-mp-2026-10-28-005-001",
        "text": "Across 1,039 tasks, structured inherited state improved repository-evidence recovery for every tested receiver and continuation for most.",
        "source_ids": [
          "source-2026-10-28-005"
        ],
        "qualification": "Six receiving models performed differently across the two tracks, showing that memory retrieval alone does not predict whether a model will honor completed work and prior commitments."
      }
    ],
    "source_ids": [
      "source-2026-10-28-005"
    ],
    "tags": [
      "AI agents",
      "handoffs",
      "repository work",
      "continuity"
    ],
    "image_url": null,
    "corrections": []
  },
  "sources": [
    {
      "source_id": "source-2026-10-28-005",
      "title": "arXiv preprint 2610.36523, original v1",
      "publisher": "arXiv",
      "url": "https://arxiv.org/abs/2610.36523v1",
      "canonical_url": "https://arxiv.org/abs/2610.36523v1",
      "source_type": "primary_research",
      "is_primary_source": true,
      "published_at": "2026-10-26T20:00:00.000-04:00",
      "accessed_at": "2026-10-28T08:24:14.982-04:00",
      "supports_claim_ids": [
        "claim-mp-2026-10-28-005-001"
      ]
    }
  ],
  "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 Handoff Has to Preserve Commitments, Not Just Facts",
    "publisher": "The Machine Press",
    "published_at": "2026-10-28T09:00:00.000-04:00",
    "canonical_url": "https://themachinepress.com/story/mp-2026-10-28-005/a-handoff-has-to-preserve-commitments-not-just-facts"
  }
}