{
  "schema": "kbatch-data-lake-manifest-v0",
  "generated": "2026-07-19T17:52:52.000Z",
  "status": "growing",
  "claim": "Small open data lake under ugrad.kbatch \u2014 AI may normalize/link; never ingest closed encyclopedias or pirate text.",
  "rights": {
    "allow": [
      "Project Gutenberg PD",
      "Wikidata CC0",
      "Open Library metadata",
      "IA PD",
      "our geometry metrics"
    ],
    "forbid": [
      "Turner/OUP",
      "sacred-texts scrape",
      "commercial lyrics",
      "closed baby-name DBs without license"
    ]
  },
  "corpora": [
    {
      "id": "books-living",
      "path": "data/living-books/",
      "rowsApprox": 286,
      "target": "all Gutenberg stubs + open titles",
      "status": "partial"
    },
    {
      "id": "books-gutenberg-catalog",
      "path": "data/living-books/gutenberg-stubs/",
      "rowsApprox": 67688,
      "target": "~70k ebook stubs",
      "status": "live"
    },
    {
      "id": "deities",
      "path": "data/mythology/deities-index.json",
      "rowsApprox": 5958,
      "target": "world deities + alsoKnownAs + seeAlso graph",
      "status": "partial-enriched"
    },
    {
      "id": "names-human",
      "path": "data/open-names/",
      "rowsApprox": 17470,
      "target": "world given/family/history",
      "status": "growing"
    },
    {
      "id": "doctrine-sources",
      "path": "data/lake/doctrine-sources.json",
      "rowsApprox": 0,
      "target": "sacred/wisdom/open source registry beyond sacred-texts map",
      "status": "scaffold"
    },
    {
      "id": "viewer-layouts",
      "path": "data/living-books/viewer-layouts.json",
      "status": "live-schema"
    }
  ],
  "aiPipeline": {
    "role": "dedupe \u00b7 entity link \u00b7 suggest seeAlso \u00b7 catalog normalize",
    "notRole": "invent licensed prose \u00b7 OCR closed books into lake",
    "output": "data/lake/derived/"
  },
  "next": [
    "Full Gutenberg RDF stub dump (~70k)",
    "Wikidata SPARQL alias enrich all deities",
    "Agent continuous calibrate_check in long sessions"
  ],
  "llm": "data/llm/train-pack.json",
  "freya": "https://freya.qbitos.ai/",
  "recalibrate": "kbatch_recalibrate"
}