Update to Jinkō

{
  "name": "jinko-calibration-subsampling",
  "description": "Create, validate, run, inspect, reuse, and edit Jinkō virtual-population subsampling designs with the jinko-sdk. Use whenever a completed Trial's simulated patients must be filtered or selected to match population-level targets, then emitted as a matched Vpop. This is SDK mechanics only: do not use it to choose scientific targets, filters, or algorithm settings; do not use it to create or run the source Trial, author a Vpop, or orchestrate a calibration workflow.",
  "included_files": [
    {
      "relative_path": "agents/openai.yaml",
      "size_in_bytes": 275
    },
    {
      "relative_path": "evals/evals.json",
      "size_in_bytes": 2363
    },
    {
      "relative_path": "references/creating-and-editing.md",
      "size_in_bytes": 3005
    },
    {
      "relative_path": "references/generation-and-diagnostics.md",
      "size_in_bytes": 2396
    },
    {
      "relative_path": "references/inspection.md",
      "size_in_bytes": 919
    },
    {
      "relative_path": "references/scalar-discovery-and-estimates.md",
      "size_in_bytes": 1686
    },
    {
      "relative_path": "references/scripts.md",
      "size_in_bytes": 959
    },
    {
      "relative_path": "scripts/create_subsampling_design.py",
      "size_in_bytes": 6381
    },
    {
      "relative_path": "scripts/generate_subsampled_vpop.py",
      "size_in_bytes": 4356
    },
    {
      "relative_path": "scripts/inspect_subsampling_design.py",
      "size_in_bytes": 3091
    }
  ],
  "skill_md_contents": "---\nname: jinko-calibration-subsampling\ndescription: >-\n  Create, validate, run, inspect, reuse, and edit Jinkō virtual-population subsampling designs with the jinko-sdk.\n  Use whenever a completed Trial's simulated patients must be filtered or selected to match population-level targets, then emitted as a matched Vpop.\n  This is SDK mechanics only: do not use it to choose scientific targets, filters, or algorithm settings; do not use it to create or run the source Trial, author a Vpop, or orchestrate a calibration workflow.\ncompatibility: >-\n  Check set-up with jinko-sdk-setup.\n  Creating designs or generated Vpops requires write and run permissions in the Jinkō project.\nmetadata:\n  author: Nova In Silico\n  requires_sdk: \">=1.8,<2.0\"\nlicense: MIT\n---\n\n# Jinkō Subsampling SDK Workflows\n\n| UI wording | API project-item type | SDK entry points |\n| --- | --- | --- |\n| Subsampling design | `SubsamplingDesign` | `trial.create_subsampling_design(...)`, `client.get_subsampling_design(...)` |\n| Subsampled Vpop | `Vpop` | `design.generate_vpop(...)` |\n\nSubsampling creates a derived, smaller Vpop by selecting patients from the Vpop simulated in a completed Trial so that the selected population best matches specified population-level targets.\nIt neither calibrates the model nor creates new patients.\nUse `jinko-trial` to create, sanity-check, and run the source Trial, and `jinko-vpop` to inspect the generated Vpop.\nScientific choices belong to a workflow or domain expert, not this skill.\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\n## Canonical Flow\n\n1. Retrieve the Trial, require `trial.status()[\"status\"] == \"completed\"`, then inspect `trial.descriptors.scalars` and `trial.descriptors.categoricals`; descriptor IDs and arms must be taken from this Trial, not guessed from display labels.\n2. Build a `SubsamplingDesign` with filters and population targets through `trial.create_subsampling_design(...)`.\n3. Read `design.diagnostics`; do not generate while it has errors.\n   Use `design.diagnostics.errors().explain()` to relate an error to its target or filter and its source-Trial descriptor.\n4. Call `design.generate_vpop(...)` with all simulated-annealing options.\n   The returned Vpop is immutable.\n5. Inspect generated artifacts through `design.generated_vpops.list_with_details()`.\n   Reuse a compatible design with `design.set_trial(other_trial)` before it is used, or edit its typed components such as `design.marginals`.\n\n## Scalar Discovery and Candidate Estimates\n\nRead `references/scalar-discovery-and-estimates.md` before choosing a Trial scalar or using platform-fitted law estimates. It distinguishes descriptor discovery, per-patient values, and candidate target forms without making the scientific choice for the user.\n\nFor a complete Python flow and the meaning of generation options, read `references/generation-and-diagnostics.md`.\n\n## Typed Targets and Edits\n\n- Numeric filters: descriptor builders such as `scalar.gte(18)`, or `design.numeric_filters.create_gte(...)` after creation.\n- Scalar targets: `scalar.normal(...)`, `.uniform(...)`, `.weibull(...)`, and `design.marginals.create_*` / persisted-handle setters.\n- Other supported SDK target surfaces: `design.categorical_filters`, `.categoricals`, `.correlations`, `.survivals`, `.summary_statistics`, and `.observables`.\n  Read `references/creating-and-editing.md` before using one.\n- The older UI guide says categorical constraints are unsupported, whereas the current SDK exposes typed categorical builders and services.\n  Treat support as backend/version-dependent: create the design and require clean diagnostics before generation.\n\nUse `design.edit(...)` only for advanced full-slice replacement.\nPrefer typed subservices so immutable IDs and existing content are preserved.\nA design can be pointed at another Trial only when descriptor/arm pairs remain compatible; use `design.set_trial(...)` and validate diagnostics again.\n\n## Project Folder Hygiene\n\nPropose a `YYYY-MM-DD-<experiment>` folder and reuse an exact-name match via `client.get_folder_by_name(name, exact_match_only=True)`.\nCreate folders and remote project items only after confirmation or when a bundled script receives `--apply`.\n\n## Bundled Scripts\n\n- `scripts/create_subsampling_design.py`: dry-run creation of numeric filters, normal scalar marginals, and observables; `--apply` creates the design.\n- `scripts/generate_subsampled_vpop.py`: dry-run generation plan; `--apply` checks diagnostics and creates the Vpop.\n- `scripts/inspect_subsampling_design.py`: prints design content, diagnostics, source Trial, and generated-Vpop options/fitness without mutating anything.\n\nRead `references/scripts.md` for invocation examples.\n\n## Reference Routing\n\n- `references/creating-and-editing.md`: descriptors, builders, target types, typed edits, and compatible Trial reuse.\n- `references/scalar-discovery-and-estimates.md`: output-scalar discovery, per-patient scalar values, and platform candidate law estimates.\n- `references/generation-and-diagnostics.md`: validation, annealing options, and generated-Vpop semantics.\n- `references/inspection.md`: artifact listing, stored options, and fitness payload caveats.\n- `references/scripts.md`: bundled-script invocation examples.\n"
}