{
  "$schema": "https://themachinepress.com/schemas/story-v1.schema.json",
  "schema_version": "1.0.0",
  "document_type": "machine_press_story",
  "story": {
    "story_id": "mp-2026-08-16-007",
    "source_story_id": "tmp-story-transactional-prompt-corpus",
    "edition_id": "mp-2026-08-16-morning-0038",
    "edition_url": "https://themachinepress.com/edition/2026-08-16",
    "position": 7,
    "story_type": "dispatch",
    "section": "research",
    "editorial_classification": "editorial",
    "headline": "Fifty-Seven Thousand Prompts Became a Dataset",
    "slug": "fifty-seven-thousand-prompts-became-a-dataset",
    "dek": "Researchers collected transactional prompts embedded in software and mapped their formal, semantic and usage patterns.",
    "summary": "Researchers collected transactional prompts embedded in software and mapped their formal, semantic and usage patterns.",
    "body_text": "The corpus contains 57,500 unique prompts gathered from GitHub, focused on reproducible instructions integrated into code rather than one-off chat messages. A structured ontology records properties, formal components and semantics, revealing a Zipf-like mix of dominant patterns and a long tail across languages, domains, tasks and modalities. The team released both the dataset and an exploration interface, along with an error analysis of its automated annotations.",
    "why_it_matters": "Researchers collected transactional prompts embedded in software and mapped their formal, semantic and usage patterns.",
    "limitations": [],
    "importance": 8,
    "canonical_url": "https://themachinepress.com/story/mp-2026-08-16-007/fifty-seven-thousand-prompts-became-a-dataset",
    "json_url": "https://themachinepress.com/story/mp-2026-08-16-007.json",
    "first_published_at": "2026-08-16T09:00:00.000-04:00",
    "modified_at": "2026-08-16T15:43:36.912-04:00",
    "content_status": "new",
    "is_carryover": false,
    "carryover_reason": null,
    "key_claims": [
      {
        "claim_id": "claim-mp-2026-08-16-007-001",
        "text": "Researchers collected transactional prompts embedded in software and mapped their formal, semantic and usage patterns.",
        "source_ids": [
          "source-2026-08-16-007"
        ],
        "qualification": null
      }
    ],
    "source_ids": [
      "source-2026-08-16-007"
    ],
    "tags": [
      "prompts",
      "GitHub",
      "dataset",
      "software"
    ],
    "image_url": null,
    "corrections": []
  },
  "sources": [
    {
      "source_id": "source-2026-08-16-007",
      "title": "arXiv preprint 2608.12905",
      "publisher": "arXiv",
      "url": "https://arxiv.org/abs/2608.12905",
      "canonical_url": "https://arxiv.org/abs/2608.12905",
      "source_type": "primary_research",
      "is_primary_source": true,
      "published_at": "2026-08-13T03:43:34.000-04:00",
      "accessed_at": "2026-08-16T15:41:08.010-04:00",
      "supports_claim_ids": [
        "claim-mp-2026-08-16-007-001"
      ]
    }
  ],
  "corrections": [],
  "publisher": {
    "name": "The Machine Press",
    "url": "https://themachinepress.com",
    "description": "A daily newspaper for the age of artificial intelligence."
  },
  "cite_this_report": {
    "title": "Fifty-Seven Thousand Prompts Became a Dataset",
    "publisher": "The Machine Press",
    "published_at": "2026-08-16T09:00:00.000-04:00",
    "canonical_url": "https://themachinepress.com/story/mp-2026-08-16-007/fifty-seven-thousand-prompts-became-a-dataset"
  }
}