service-page-writer
Skill0 starsChanged 5 months agoMITView on GitHub
Writes high-converting service pages for local businesses (plumbers, dentists, restaurants, contractors) using a proven page skeleton, local copy rules, and schema markup guidance. Use when the user needs a page for a specific service, says "write a service page", "we need a page for [service]", or wants SEO landing copy for a local service.
In this repository11 other files it gives its agents.
- cold-email-personalizerSkill
- google-business-profile-optimizerSkill
- invoice-chaserSkill
- landing-page-critiqueSkill
- local-seo-auditSkill
- midwest-modeSkill
- missed-call-textbackSkill
- proposal-writerSkill
- review-responderSkill
- viral-hooksSkill
- website-conversion-auditSkill
The file
We split this skill into the rules it gives an agent
5 rules. About 0 are common, 4 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈4
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 12
Never write from the service name alone.
- Very specific: line 56
If the site has no sitewide LocalBusiness schema, include it nested in provider with name, address, phone, URL, hours, geo.
- Distinctive: line 65
Don't write a word before intake.
- Distinctive: line 66
Don't keyword-stuff city names.
- Distinctive: line 67
Don't invent reviews, stats, or guarantees.
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
Service Page Writer
A service page has two readers: a stressed human deciding who to call, and a search engine deciding who to show. Write for the human in a structure the engine understands.
Step 1 — Intake
Never write from the service name alone. Collect:
- The service, and what jobs it actually includes/excludes
- Business name, city, service area
- Price signal the owner will commit to ("from $X", "typical range $X–Y", or "free quote" if truly nothing else)
- 2–3 real differentiators (years, jobs count, response time, guarantee, licenses) — push past adjectives: "what can you say that a competitor can't copy-paste?"
- 2–3 real reviews mentioning this service (names + towns)
- Photos available of this work
- The 5 questions customers actually ask before booking this service
- Emergency/after-hours availability, financing, brands used
Step 2 — Build on the skeleton
Eleven blocks, in order. Every block earns its place; skip a block only if the intake gave nothing real for it.
- H1:
[Service] in [City]— plus the strongest differentiator if it fits naturally. - Subhead: outcome + trust in one line. "Fixed right the first time — 1,400+ water heaters since 2009."
- CTA block: clickable phone + one button ("Get a free quote"). Repeats after blocks 6 and 11.
- The situation: 2–3 sentences in the customer's shoes — the problem as they'd describe it, not as the trade describes it. ("No hot water this morning" — not "sediment accumulation reduces heating efficiency.")
- What's included: bulleted, concrete. What happens on the visit, what's covered, what they leave with.
- Price transparency: the range and what moves it. Owners resist this; make the case — the customer will get a number somewhere, and the site that gives it earns the call. Even "most jobs land between $X–$Y; here's what pushes it up or down" beats silence.
- Why us: the intake differentiators, stated as facts with numbers. Three maximum.
- Process: 3–4 steps from call to done. Kills the fear of the unknown.
- Reviews: 2–3, this service specifically, name + town.
- Service area: the towns, written as a sentence, not a keyword pile.
- FAQ: 5–7 of the real pre-booking questions. Each answer is complete enough to stand alone (these are schema fodder and objection handlers).
A fully worked example page (plumber, water heater replacement) is in references/template.md.
Step 3 — Copy rules
- Reading level ~8th grade. Short sentences. No trade jargon without a plain-word gloss.
- "You" outnumbers "we." Count if unsure.
- Specifics beat adjectives everywhere: "background-checked, in your driveway inside 2 hours" not "reliable and professional."
- Banned: "Welcome to," "we pride ourselves," "your satisfaction is our priority," "look no further."
- The page must answer cost, time, and "will they show up" — those are the three real questions.
- Multi-city businesses: one page per service; city pages only with city-specific proof (see the local-seo-audit skill).
Step 4 — Schema markup
Emit JSON-LD with the page, in a code block, with a note on where it goes (one <script type="application/ld+json"> in the head):
- Service schema:
serviceType,provider(LocalBusiness with NAP exactly matching the site footer),areaServed. - FAQPage schema from the FAQ block — questions and answers verbatim from the page (never schema-only content; that's a spam signal).
- If the site has no sitewide LocalBusiness schema, include it nested in
providerwith name, address, phone, URL, hours, geo. - Tell the user to validate at Google's Rich Results Test after publishing.
Step 5 — Deliver
Full page in markdown with block labels as comments, then the schema block, then a 3-line publish checklist (title tag, meta description — both provided — and internal links from the homepage and related service pages).
What not to do
- Don't write a word before intake. A page without price signals, real reviews, and real differentiators is furniture.
- Don't keyword-stuff city names. Once in the H1, once or twice in body, once in the service-area block — done.
- Don't invent reviews, stats, or guarantees. If intake gave none, write the page without that block and tell the user what it's missing.
Built by Reef — AI systems for small business → https://buildwithreef.com
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.

