shopify-seo-brand
Skill0 starsChanged 2 months agoMITView on GitHub
Create or update the brand fact sheet that every other Claude Shopify SEO skill checks copy against. Use for: set up my brand, add my shop, change my facts, update prices, my voice, what I am allowed to claim.
In this repository7 other files it gives its agents.
- shopify-seo-checkSkill
- shopify-seo-fixSkill
- shopify-seo-geoSkill
- shopify-seo-learnSkill
- shopify-seo-publishSkill
- shopify-seo-schemaSkill
- shopify-seoSkill
The file
We split this skill into the rules it gives an agent
8 rules. About 0 are common, 5 are distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Read the live store first, then ask.
- Distinctive: line 14
If the Shopify MCP is connected, pull the shop name, currency, product types, live price ranges and size options yourself.
- Very specific: line 27
Do not ask the owner to describe their own voice, they will describe how they wish it sounded.
- Very specific: line 28
Read four or five of their existing descriptions or posts and propose a voice block back to them for correction.
- Distinctive: line 36
Never fill a field to look complete.
- Distinctive: line 44
Keep it in the owner's own working folder.
- Distinctive: line 47
If you are working under a services agreement, confidentiality of client data is usually an express term of it.
- Very specific: line 52
Write brand/<slug>.json following the shape of brand/shibumi.json, then show the owner the facts you recorded and ask them to correct …
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Brand
The fact sheet is the whole point. Everything else checks against it, so it has to be true.
Building one
Read the live store first, then ask. Do not ask for what you can verify.
If the Shopify MCP is connected, pull the shop name, currency, product types, live price ranges and size options yourself. Then ask the owner only for what a machine cannot see:
- What they make, in their own words, in one sentence
- Lead times, and any paid express option
- Made to measure or bespoke, and whether they are different things here
- Anything they design or make themselves, which is usually their strongest claim
- Whether any real ratings or reviews exist, and where
- Phrases they never want used
- Anything seasonal with a hard date, such as an ordering cutoff
Voice
Do not ask the owner to describe their own voice, they will describe how they wish it sounded.
Read four or five of their existing descriptions or posts and propose a voice block back to them
for correction. Note what they actually do, including habits they may not have noticed.
Rules
- Verify prices against the live store and record the date checked. Prices go stale and stale prices in copy are a real problem.
- Where a fact holds for most items but not all, say so in the value itself rather than rounding it up.
- Never fill a field to look complete. An absent field is honest. A guessed one contaminates everything downstream, because every other skill treats this file as true.
Confidentiality
A brand profile holds a client's prices, lead times, margins-adjacent detail and often a candid list of what is wrong with their site. It is their commercial data, not yours.
Keep it in the owner's own working folder. Do not commit it to a public repository, do not paste it into an issue, and do not ship it as an example. Where a plugin needs a sample, write a fictional one.
If you are working under a services agreement, confidentiality of client data is usually an express term of it. Treat this as a contractual obligation rather than a preference.
Output
Write brand/<slug>.json following the shape of brand/shibumi.json, then show the owner the facts
you recorded and ask them to correct anything wrong. Do not move on until they have read it.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

