Update to Jinkō

{
  "name": "jinko-task-extract-data-table",
  "description": "Extract or digitize reported biomedical values from papers, figures, tables, supplements, images, or web sources into traceable CSV/Markdown, optionally as a calibration-ready Jinkō data table. Use when numeric evidence must be transcribed, normalized, unit-converted, or bound to model observables. Do not use for literature discovery, evidence synthesis, or inventing values absent from the source.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 271
    }
  ],
  "skill_md_contents": "---\nname: jinko-task-extract-data-table\ndescription: >-\n  Extract or digitize reported biomedical values from papers, figures, tables,\n  supplements, images, or web sources into traceable CSV/Markdown, optionally as\n  a calibration-ready Jinkō data table. Use when numeric evidence must be\n  transcribed, normalized, unit-converted, or bound to model observables. Do not\n  use for literature discovery, evidence synthesis, or inventing values absent\n  from the source.\ncompatibility: >-\n  Jinkō upload requires jinko-sdk-setup and project write access. Extraction from\n  images or PDFs requires a suitable reader, OCR, or digitization tool.\nmetadata:\n  author: Nova In Silico\n  requires_sdk: \">=1.8,<2.0\"\nlicense: MIT\n---\n\n# Extract Data Table\n\n> **PREREQUISITE:** This skill needs an initialized `jinko-sdk` connection and an\n> SDK satisfying its `metadata.requires_sdk` range. Run the `jinko-sdk-setup` skill\n> (`../jinko-sdk-setup/SKILL.md`) and proceed only once its check passes. If that\n> skill is not found, install it from `novainsilico/jinko-skills`.\n\nPreserve what the source reports. Keep estimated, transformed, and directly\ntranscribed values distinguishable.\n\n## Inputs\n\nRequire the source artifact and the requested series. For a Jinkō-ready output,\nalso require the target `obsId` mapping, model units, and any scenario/arm scope.\nAsk for missing mappings rather than guessing them.\n\n## Workflow\n\n1. Locate each requested series and record its citation plus page, table, figure,\n   panel, or supplement. Prefer machine-readable tables over OCR and OCR over\n   graphical digitization. If no suitable extraction tool is available, request\n   a tabular source instead of estimating visually.\n2. Extract only reported values. For graphical digitization, retain the raw\n   digitized points and identify them as estimates. Do not fit, smooth, aggregate,\n   or impute unless explicitly requested; record any such transformation.\n3. Preserve the reported statistic. Do not interchange raw values, means,\n   medians, SD, SE, confidence intervals, IQR, or min/max. Convert units only when\n   the source and target units are known, recording the formula and original\n   values. Express Jinkō time values as ISO-8601 durations.\n4. For general extraction, emit a readable CSV or Markdown table with series,\n   time/condition, value or bounds, unit, and source locator.\n5. For Jinkō output, follow the row schema owned by `jinko-data-table`. Use point\n   rows for reported point values and range rows only for reported lower/upper\n   bounds. Set `obsId`, `armScope`, `unit`, and `experimentRef` explicitly.\n6. Run the `jinko-data-table` creation script in dry-run mode with the expected\n   `--allowed-obs-id` values, `--require-unit`, and `--require-experiment-ref`.\n   On approval, add `--require-fitness --apply`. The script owns row/schema\n   checks, upload, and the server `validForFitnessFunction` gate.\n\n## Return\n\nReturn the extracted file and a compact report containing:\n\n- source locator and extraction method for each series;\n- original statistic, units, and any transformations or conversions;\n- assumptions, unreadable values, and digitization uncertainty;\n- for Jinkō output, data-table SID, URL, observable mapping, and confirmed\n  `validForFitnessFunction: True`.\n\nIf the server does not explicitly report fitness compatibility as true, return\nthe table as not calibration-ready. Never silently replace or manufacture values\nto make a table pass validation.\n"
}