01
Step 1
Collect the URL list in scope, each URL's current title and description fields, the rendered page headings, and the search intent the merchant states for each page.
Back to topic
Spark playbook scenarios · Spark: AI Store Assistant
Audit on-page SEO metadata and content issues
A merchant wants a single pass over the storefront that says which pages are under-optimised before any content work begins. In this case, the immediate task is to audit on-page SEO metadata and content issues. The workflow should produce a row per URL recording what is missing, duplicated or off-intent, with a replacement within the length rules used.
Topic
SEO Metadata and Content Audits
Target keyword
shopify seo audit tool
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 audit on-page SEO metadata and content 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: [the URL list in scope, current title and description fields per URL, rendered page headings, and the search intent stated per page]. Flag per URL the values that are missing from the record, duplicated across URLs, or mismatched with the stated intent, classifying missing as a confirmed fact and mismatch as a proposed revision, and mark URLs whose copy or intent is not supplied as unresolved.
Return a URL-level audit with proposed revisions kept inside the stated length rules 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
Flag per URL the values that are missing from the record, duplicated across URLs, or mismatched with the stated intent, classifying missing as a confirmed fact and mismatch as a proposed revision, and mark URLs whose copy or intent is not supplied as unresolved.
03
Step 3
Deliver the URL-level audit with a proposed revision inside the stated length rules for every confirmed finding, then check each row against the supplied fields so nothing is reported as missing that the record shows.
FAQ
Common questions when running this scenario
Cover decision rules, frequent misunderstandings, and required inputs so the team aligns on prerequisites before execution.
How are facts and judgements kept apart in the audit?
A missing field is confirmed straight from the record and reported as an error, while a weak or off-intent value is a judgement and appears as a proposed revision, not a defect. The two carry different labels in the same table. This keeps the audit's factual claims defensible while still giving the merchant actionable rewrites.
Which URLs end up marked unresolved?
URLs whose visible copy or stated search intent was not supplied, since without one of them the audit can list what to check but cannot confirm a problem. They are listed separately from findings, so they never pad the issue count. They move into the audit once the merchant supplies the missing material.
What length rules do the replacements follow?
The title and description rules the merchant states; where no rule is supplied the constraint is recorded as missing and the draft is written to the platform's own display limits rather than an invented count. Each replacement notes which rule it followed. Drafts are never accepted as final while the length constraint is still unknown.
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.