The merchant wants problems surfaced before customers report them, but the team never agreed what counts as an issue. In this case, the immediate task is to detect and prioritize evidenced store issues. The workflow should produce a prioritized issue list with the source of each finding, grounded in the merchant's own checks and signals.
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 detect and prioritize evidenced store issues.
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: [each registered check with its source and fields, expected range, lookback history, allowed source freshness, and the merchant objective it protects]. Run every check against its stated condition and sort results into confirmed issues, stale-source signals where the feed did not refresh inside its allowed window, and single-source anomalies awaiting corroboration, then rank the confirmed issues by their effect on the stated objective.
Return the prioritized issue list with the source of each finding verified against the supplied records 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.
Register each check with the source and fields it reads, the expected range separating normal movement from a fault, the lookback history that justifies that range, the freshness the source must meet, and the merchant objective the check protects.
Run every check against its stated condition and sort results into confirmed issues, stale-source signals where the feed did not refresh inside its allowed window, and single-source anomalies awaiting corroboration, then rank the confirmed issues by their effect on the stated objective.
Deliver the prioritized issue list naming per finding the check, source, quoted evidence, expected range and reconfirmation step, then reconcile against the check inventory so every registered check reports a result including the ones that passed.
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
Confirm the source refreshed inside its allowed freshness window before accepting the signal; a late feed traces the same movement as a real drop, so such signals are logged as data gaps and rechecked once the source is current.
No. Confirmed issues are worked in ranked order and one change at a time, because bundling fixes leaves the reconfirmation check unable to show which intervention actually produced the recovery.
A stated expected range, enough history to describe normal movement for that signal, a freshness rule for its source, and a second record able to corroborate it; a check missing any of these stays advisory rather than a basis for action.
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.