01
Step 1
Collect the affected product URLs and handles, the variant selected in each report, the device and browser reported, whether the shopper was signed in, and any console, network or order evidence already captured.
Back to topic
Spark playbook scenarios · Spark: AI Store Assistant
Diagnose a failing add-to-cart interaction
Shoppers report the button does nothing on specific product pages, and the affected URLs, variants, devices and error evidence are on hand. In this case, the immediate task is to diagnose a failing add-to-cart interaction. The workflow should separate stock or script faults from a poorly placed control and give a retest checklist.
Topic
Add to Cart Issues and Optimization
Target keyword
shopify add to cart button not working
Copyable AI prompt
A prompt you can drop directly into Spark
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.
ai-prompt.txt
Act as a Shopify operations specialist.
Help me diagnose a failing add-to-cart interaction.
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 affected product URLs and handles, the variant selected in each report, the reported device and browser, the shopper's signed-in state, and any captured console, network or order evidence]. Reproduce the interaction on the recorded variant and device where access allows and classify each case by evidence, as an availability or selling-plan block when the variant is unsellable or requires a plan, as a script or network failure when the request errors or never leaves the page, and as a placement or confirmation problem only when the request succeeds and the cart updates but the shopper saw no feedback.
Return an add-to-cart diagnosis and verification checklist, with each recorded case retested 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.How to solve
A recommended resolution workflow
Split the work into input, judgment, and deliverable phases. Verify each step's output before moving to the next.
02
Step 2
Reproduce the interaction on the recorded variant and device where access allows and classify each case by evidence, as an availability or selling-plan block when the variant is unsellable or requires a plan, as a script or network failure when the request errors or never leaves the page, and as a placement or confirmation problem only when the request succeeds and the cart updates but the shopper saw no feedback.
03
Step 3
Deliver the diagnosis naming the class of every reported case with the evidence behind it plus a retest checklist, then verify it by confirming that each case falls into exactly one class and that any case left unclassified is listed with what is still needed to decide.
FAQ
Common questions when running this scenario
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
How is a stock problem told apart from a script failure?
By where the request stops: if the variant is unsellable or needs a selling plan the control is disabled or the request is refused before any error appears, while a script failure shows a request that errors or never leaves the page. A request that succeeds and adds the line points instead at confirmation the theme never showed.
What if the fault cannot be reproduced?
The case stays on the checklist as unclassified, recorded with the variant, device and time reported, because intermittent script loading, network conditions and session state can fail once and still be real. It is retested after any related change rather than closed for lack of a second sighting.
Why check the cart state as well as the click?
A click can register while the cart drawer shows nothing, so the shopper experiences a dead button even though the line was added. Confirming cart contents after the click separates a control that truly failed from a confirmation the theme never displayed.
Related scenarios
Continue with adjacent or same-topic scenarios
Start with scenarios under the same topic, then expand into cross-topic matches suggested by similarity.
Next step
Want these AI scenarios embedded directly into daily operations?
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.