Step 1
Collect the repeated questions in the shopper's own wording, the approved documents and policy pages allowed as sources, and the order fields an answer may quote, such as status, carrier and dispatch date.
A merchant wants a chatbot to handle recurring pre-sale and order questions without improvising beyond approved wording. In this case, the immediate task is to answer repeated support questions from approved sources. The workflow should draft answers that stay inside the approved documents and order fields, and flag questions needing a human decision.
First replace every [bracketed] placeholder with real store scope and constraints, then add any actual records or data you have, and paste it into your preferred AI assistant or Spark agent. Spark will follow this page's workflow to produce structured output and verification steps.
Act as a Shopify operations specialist.
Help me answer repeated support questions from approved sources.
My store context is [describe the business, products and target audience].
My scope is [specify the affected products, pages, customers or reporting period]. Use these inputs: [the repeated questions in the shopper's own wording, the approved policy documents and product pages, the order fields an answer may quote and the merchant's handoff conditions].
Draft each answer so every claim traces to an approved source or an order field, keep the wording inside what those sources say, and mark for handoff any question whose answer depends on a policy exception, a decision only the merchant can make, or data no approved source holds.
Return a support knowledge base outline and grounded FAQ answers, flagging questions that need a merchant decision for the specified task.
Show the relevant evidence or before-and-after examples and explain how I can verify the result.
If essential data is missing, ask for it and mark unsupported conclusions as unknown.
Do not invent metrics, product claims or customer facts.
For any publication, message, account or store change, use only connected tools within my explicitly authorized scope and report success only after verification; otherwise provide the draft or proposed steps.Split the work into input, judgment, and deliverable phases. Verify each step's output before moving to the next.
Collect the repeated questions in the shopper's own wording, the approved documents and policy pages allowed as sources, and the order fields an answer may quote, such as status, carrier and dispatch date.
Draft each answer so every claim traces to an approved source or an order field, keep the wording inside what those sources say, and mark for handoff any question whose answer depends on a policy exception, a decision only the merchant can make, or data no approved source holds.
Deliver the knowledge base outline with the grounded answers and the handoff rule, then replay the collected questions against it and confirm each one is answered, handed off or logged as needing an approved source.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
When the answer needs order-specific data the sources do not hold, involves a policy exception, or asks the merchant to make a decision. Answering anyway commits the store to something nobody approved.
No. Anything outside the approved sources can contradict the merchant's own policy, and the shopper has no way to tell an invented detail from an approved one.
Log the exact wording and route it to the merchant for an approved response. Guessing keeps the gap invisible and leaves the same question unanswered next week.
Start with scenarios under the same topic, then expand into cross-topic matches suggested by similarity.
Ciwi Spark builds Shopify-native AI tools and agents around the workflows merchants actually run: research, ads, translation, and customer support. Stop rewriting the same prompts by hand.