data/normalized/articles_ingest_report.json

{
  "generated_at": "2026-01-28T17:19:43.408336Z",
  "out_jsonl": "data/normalized_v1_2/chunks_articles.jsonl",
  "out_manifest": "data/normalized_v1_2/manifest_articles.json",
  "total_chunks_written": 677,
  "sources_total": 29,
  "sources_skipped": [],
  "notes": [
    "Towards Data Science links may return 403 and are skipped to keep the pipeline reproducible.",
    "arXiv PDFs are ingested via pdfminer; title/author/date may be enriched later."
  ]
}