A merchant suspects several store problems at once but has never separated what can be proven from guesswork. In this case, the immediate task is to audit store health across accessible evidence. The workflow should test each suspected issue against reachable records, producing an audit of confirmed findings alongside open questions.
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 audit store health across accessible evidence.
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 store URLs and areas in scope, catalog reports, order and fulfillment exports, configuration records reachable for this pass, the suspected issues with their reasons, and a list of systems not accessible]. Test each suspected issue against the reachable records and record it as a confirmed finding only where the evidence supports it, as an open question where the evidence is missing or contradictory, and as not reviewed where the area cannot be reached, refusing to score any area in the last group.
Return an audit with confirmed findings and open questions 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.
Gather the store URLs and the areas the merchant wants examined, the catalog reports, order and fulfillment exports and configuration records reachable for this pass, the list of issues already suspected with the reason each is suspected, and an explicit note of which systems are not accessible.
Test each suspected issue against the reachable records and record it as a confirmed finding only where the evidence supports it, as an open question where the evidence is missing or contradictory, and as not reviewed where the area cannot be reached, refusing to score any area in the last group.
Deliver an audit separating confirmed findings from open questions, each finding carrying the affected records, the evidence and the next check and each open question carrying the check that would settle it, then reconcile the number of findings against the number of suspected issues entered so nothing silently disappears.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
Only when a reachable record supports it, and the finding then names the affected records, the evidence and the next check. Where evidence is missing the item stays an open question with the check that would settle it, never a finding and never silently closed.
It can list the checks those systems would require, but must not report an unreachable area as verified healthy or verified broken. Such an area is marked not reviewed with the access needed, since scoring it would invent a result the evidence cannot carry.
A score can summarise confirmed findings but does not itself say what to fix, so each finding keeps its affected records, evidence and next check. Areas marked not reviewed are excluded from the score rather than counted as passing.
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.