Generated from smart-base/skills/content/authoring-who-smart-guidelines/l3-dependency-choice.md — do not edit here.

✎ Edit this page’s source

l3-dependency-choice

Skill id: l3-dependency-choice · Package: authoring-who-smart-guidelines · Stage 5 of smart-guideline-end-to-end, task Task_DepsL3 of smart-guideline-end-to-end.bpmn. Adjudicated by the FHIR modeller and the terminologist.

The L3 half of l2-dependency-choice: the same decision row, the same reuse-first precedence, the same decision-comparison table and DecisionRequest. Only the candidates and the evidence differ.

Input and output

The contract is input.schema.json and output.schema.json.

• In: the l1Kg (stage 2), the l2Assets (every stage-4 l2-dak-authoring pass’s artifacts), stage 3’s l2Dependencies, and any candidate IGs already known.
• Out: l3Dependencies (level: L3), rows of the same dependencyRow whose components name canonicals, and indexes, the ig-artifact-ingestion index each row was read from, with its dakApi state. Stage 6 (fhir-harness l3-fhir-authoring) reads the adjudicated rows as sushi-config.yaml dependencies.

Steps

1. Candidates. Those given, plus each IG that implements a DAK stage 3 chose to reuse, adapt or extend.
2. Index each one with ig-artifact-ingestion. The index is the evidence; an IG whose root was never read has dakApi: unknown, which is not absent.
3. Match. For each L2 asset and L1 node, find the profile, value set, CQL library or PlanDefinition in the index that already implements it. The matches are the row’s evidence; what is taken is its components (reuse, adapt, extend, not-used). For each artefact type taken, record the toolchain stage 6 will use: how the candidate builds and tests its CQL and examples, found by the search in smart-base-tools §Adapting a DAK.
4. Decide, table, ask and go on, exactly as l2-dependency-choice steps 2–4.

Processes that run this skill

process	step(s) that name it
SMART Guideline end to end	5 · Choose L3 dependencies (WHO / jurisdiction IGs)