Generated from smart-base/skills/content/authoring-who-smart-guidelines/guideline-use-case.md — do not edit here.

✎ Edit this page’s source

guideline-use-case

Skill id: guideline-use-case · Package: authoring-who-smart-guidelines · Stage 0 of smart-guideline-end-to-end, task Task_DescribeUseCase of smart-guideline-end-to-end.bpmn, in the requester’s (user) lane.

The use case is the requester’s, not the agent’s. Everything downstream — which L1 sources count, which DAKs and IGs are candidates, which application is built — is chosen against it, so a use case the agent made up would bias every later stage without anyone having decided it.

Input and output

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

• In: the request, verbatim, and the jurisdiction and the guideline it adapts when the request names them.
• Out: the useCase (population, setting, jurisdiction, interventions, kind new or adaptation, the requester’s words), which stage 1 (dak-l1-library) takes as its useCase; and the numbered questions asked to get it.

Steps

1. Read the four required fields off the request. Use the requester’s words where they gave any, and keep them verbatim in words.

2. For each one the request does not say, ask a numbered question (interaction-modality), with options drawn from what is already known, a recommended one marked, and Tell me more last:

   Which jurisdiction is this guideline for?

   1. WHO (global): a new guideline, or a WHO DAK extended.
   2. A country’s adaptation of an existing WHO DAK (recommended when the request names a country).
   3. Tell me more.

3. An adaptation names what it adapts in adapts. Stages 3 and 5 then return adapt rows for it, and stages 4 and 6 show diffs.

Adjudication

None: the requester is the source. A later stage that finds the use case wrong says so to the requester, and stage 0 is redone with their answer.

Processes that run this skill

process	step(s) that name it
SMART Guideline end to end	0 · Describe the use case