Update to Jinkō

{
  "name": "jinko-task-define-param-to-calibrate",
  "description": "Classify directly valued Jinkō model inputs by evidence source and assign inputs needing calibration to explicit calibration steps. Use when the user wants to decide which parameters, categorical parameters, or species initial conditions should be calibrated and record the decision with `s::*` and `CalibIter::*` tags. Do not use for choosing datasets, estimating priors, drafting calibration plans, or running calibrations.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 287
    },
    {
      "relative_path": "scripts/apply_calibration_labels.py",
      "size_in_bytes": 9670
    }
  ],
  "skill_md_contents": "---\nname: jinko-task-define-param-to-calibrate\ndescription: >-\n  Classify directly valued Jinkō model inputs by evidence source and assign\n  inputs needing calibration to explicit calibration steps. Use when the user\n  wants to decide which parameters, categorical parameters, or species initial\n  conditions should be calibrated and record the decision with `s::*` and\n  `CalibIter::*` tags. Do not use for choosing datasets, estimating priors,\n  drafting calibration plans, or running calibrations.\ncompatibility: >-\n  Check set-up with jinko-sdk-setup. Applying labels requires model write access.\nmetadata:\n  author: Nova In Silico\n  requires_sdk: \">=1.8,<2.0\"\nlicense: MIT\n---\n\n# Define Parameters To Calibrate\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\nClassify model inputs from supplied evidence; do not infer unsupported provenance\nor invent calibration steps.\n\n## Inputs\n\nRequire:\n\n- a model SID;\n- documentation or other evidence for the current input values;\n- an explicit ordered list of calibration step identifiers and each step's biological\n  scope;\n- any user overrides and whether existing labels may be replaced.\n\n## Labels\n\nAssign at most one source tag per eligible input:\n\n- `s::knowledge`: supported by literature, expert knowledge, or a reference\n  model;\n- `s::arbitrary`: deliberately fixed without an evidence-derived value;\n- `s::to-calibrate`: insufficiently informed and intended for calibration;\n- `s::calibrated`: preserve when present; never assign in this task.\n\nAssign `CalibIter::<step>` only with `s::to-calibrate`, using an explicitly\nprovided step whose scope covers the input. Absence means the input is not\nassigned to calibration. Leave uncertain inputs unchanged and report them.\n\n## Workflow\n\n1. Use `jinko-model` to inspect parameters, categorical parameters, and species\n   initial conditions. Exclude derived formulas, technical infrastructure, and\n   non-input components.\n2. Preserve existing `s::*` and `CalibIter::*` tags unless relabeling was\n   requested. For each remaining input, classify its source from the evidence.\n3. Map every `s::to-calibrate` input to the first supplied step whose scope fully\n   covers its biological role. If no unique step qualifies, leave it unchanged\n   and add it to `todo`.\n4. Write the proposed mutations as JSON and run\n   `scripts/apply_calibration_labels.py` in dry-run mode, then with `--apply`\n   after review. The script validates source/step consistency, duplicate\n   assignments, component kinds, existing-label conflicts, and allowed model\n   mutations before applying one component batch.\n5. Re-fetch the model and return the new revision and snapshot with a compact\n   report: assigned and preserved labels, counts by source and step, and `todo`\n   entries with reasons.\n\nThe mutation plan has this shape:\n\n```json\n{\n  \"in_scope_steps\": [\"2\", \"3\"],\n  \"assignments\": [\n    {\"component_id\": \"k_elim\", \"source\": \"to-calibrate\", \"calibration_step\": \"2\"},\n    {\"component_id\": \"body_weight\", \"source\": \"knowledge\"}\n  ]\n}\n```\n\nDo not change values, units, descriptions, equations, structure, or unrelated\ntags. `todo` items are reported, not encoded as placeholder tags.\n"
}