SuperAPI AI Directory

question-intake

Office & productivity

View repository

Updated 11-03 10:02

About this skill

Question Intake

Purpose

Structure the question with the context that changes the answer: store, cluster, department, role, date, SKU(s).

When to use

Start of every assist run - any policy/product/promotion question from the floor.

Inputs

The question as asked (chat/voice transcript); store profile (id, format, cluster); asking role. Emits the entry hop of rtl.store-associate-assist.v1 (schema in contracts/).

Steps

  1. Capture the question verbatim; extract topic hints, SKU/GTIN(s) if named, compare list if two products.
  2. Attach store_context (store id, cluster, department) and asked_on date - promo eligibility depends on all three.
  3. Validate against the contract; write question.json.

Output

question.json - the entry hop.

Grounding requirements

The question text stays verbatim; store context cites the store master.

Constraints

  • No answering here; resolution belongs to policy-retrieve.
  • Never guess a GTIN - ask the associate to scan if ambiguous.

Escalation / uncertainty

Missing store context blocks promo checks - ask for the store id.