Plan an e-commerce category / collection page that ranks and merchandises well.
SkillScore82Maintained
Automatic
◆Marketing & Content pack
Last verified 8/14/2026
What it changes
Passed 18 of 22 test cases with the skill, and 16 of 22 without it.
Without skill73%
With skill82%
Checks it passed that the model alone missed
Tags inferred target keywords with confirmation markers.
Flags inferred keyword search terms with confirmation markers.
2 fixed16 passed either way4 still failing0 broke
Pass-rate lift
+9 pp
Tokens
+19%
Agent turns
0%
Response time
-6%
Model gemini-3.6-flashMeasured 8/14/2026v118 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 2
Prompt
I need a category page specification for our line of ergonomic office chairs. Please put together a brief outlining commercial intent and keyword targets, top-of-page copy and H1 tag, default product sorting and inventory rules, filter crawlability controls, internal linking architecture, trust elements, technical SEO metadata, and a bottom-of-page buying guide or FAQ section.
What was checked, without and with the skill
Checked
Without skill
With skill
Flags inferred keyword search terms with confirmation markers.
Missed
Passed
Specifies noindex directives for multi-filter URL combinations to control crawl space.
Passed
Passed
case-03 · gemini-3.6-flash · 8/14/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md57 linessingle file
Category Page Brief Skill
Category pages are the workhorses of e-commerce SEO and discovery — they rank for high-intent "buy" searches
and decide what a browsing shopper sees first. A good one balances search relevance (the right keywords,
copy, and structure) with merchandising (the right products, order, and filters). This skill briefs that
page so it earns traffic and converts it.
Working from a brief
Given "a category page for women's running shoes", produce the full brief anyway — infer the search intent,
likely keywords, sensible filters, and merchandising logic, marking inferences. Don't invent search volumes or
inventory; flag them to confirm. Never withhold for missing detail.
Required Inputs
Ask for these only if they aren't already provided (else infer and label):
The category — what it covers, and where it sits in the catalogue hierarchy.
The shopper & intent — who lands here and what they're trying to do (browse vs. specific need).
Inventory & attributes — roughly what products/variants exist and their key attributes (for filters).
SEO context — target keywords if known, and the platform (Shopify, Magento, custom).
Output Format
Category Page Brief: [category]
Search intent & keywords — the primary head term, secondary/long-tail terms, and the intent (commercial). Note any to validate with real volume.
Page H1 & intro copy — the H1 and a short (skimmable, keyword-aware) intro that helps shoppers and SEO — placed so it doesn't push products below the fold.
Merchandising & default sort — what shows first (bestsellers, new, margin, seasonal) and the rule behind it; promoted/pinned products; out-of-stock handling.
Filters & facets — the filters shoppers need (price, size, colour, brand, attribute), and which should be crawlable vs. noindex to avoid thin/duplicate pages.
Internal linking — links to subcategories, related categories, and key products; breadcrumb structure.
Trust & conversion elements — reviews/ratings on cards, badges, shipping/returns reassurance, a strong category image.
SEO / technical notes — title tag & meta description, canonical strategy for filtered URLs, pagination, and structured data.
Supporting content — an optional bottom-of-page FAQ/buying-guide block for long-tail terms.
Mark inferred keywords/inventory (confirm).
Quality Checks
Target keyword and search intent are explicit and reflected in H1, intro, and title tag
Write a human changelog from the REAL commit history — read the actual commit range via the GitHub connector, not a template. Use when asked to write the changelog for this release, what changed since the last tag, draft release notes from my commits, or summarise this range for users in Cowork. Reads commits/PRs between two refs via the GitHub connector, groups them into user-facing changes (features / fixes / breaking), translates commit-speak into human benefit, and produces a changelog artif