service
Skill1 starsChanged 6 months agoApache-2.0View on GitHub
Generate a service page (commercial-intent landing page) for a chosen managed site. Footprint-aware (single-location / multi-location / service-area / national-online), enforces anti-doorway-page rules, applies real NAP per Section 4.3 of SEO_GUIDE.md, includes mandatory conversion elements per Section 4.4, validates schema. Use when the user types `/service` or asks for a service page.
11 other projects use a version of this.
Adopted by
Most recently taken up by
- artivilla/agent-skillstheir own version, first seen Nov 8, 2026
- MohammadShehadeh/agent-skillstheir own version, first seen Nov 1, 2026
- vijaysuperpositionist/skillstheir own version, first seen Nov 1, 2026
- wheslleyrimar/BuildCLI-Agentstheir own version, first seen Oct 22, 2026
- The-Utopia-Studio/skillstheir own version, first seen Oct 13, 2026
- ch4570/vulporatheir own version, first seen Oct 11, 2026
Used with
Projects with it also use status.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
AnthropicSkillstatus
Case status summary by audience — client-facing (plain language), internal (for the professor), or court-ready (formal caption format per local rules). Same facts, different framing and depth. Use when a student needs to update the client, brief the professor, or prepare a court status report.
Also in 3 projects that use this
Other versions11 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 11
- spotify/portal-ai-plugins★ 2.8ksame name
- crisandrews/ClawCode★ 64same name
- landaozhu/skills★ 27same name
- maxwellsdm1867/wheeler★ 12same name
- The-Utopia-Studio/skills★ 6same name
- gabriellst/codm★ 4same name
- wheslleyrimar/BuildCLI-Agents★ 3same name
- MohammadShehadeh/agent-skills★ 2same name
- ch4570/vulpora★ 1same name
- artivilla/agent-skills★ 0same name
1 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository20 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- alternativeSkill
- auditSkill
- blogSkill
- broken-backlinksSkill
- case-studySkill
- clusterSkill
- comparisonSkill
- competitor-backlinksSkill
- glossarySkill
- haroSkill
- integrationsSkill
- lovable-deploySkill
- pricingSkill
- programmatic-batchSkill
- refreshSkill
- semrush-baselineSkill
- serp-featuresSkill
- triageSkill
- wireframeSkill
The file
Show its rules and text
We split this skill into the rules it gives an agent
17 rules. 0 are common, 5 are distinctive, and 12 are very specific to this project.
0
Common
Most of the 8 edited versions we read keep these.
5
Distinctive
This author's own choices. Worth a look.
12
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 12
If sites/[site]/site-info.md → Business → Service business: false, the skill refuses to run and points the user to: "Use /blog for …
- Very specific: line 34
If the invocation contains --cron, this skill runs in cron mode (no user available).
- Very specific: line 40
Defaults: when a choice would normally be asked, default to site-info.md / goals.md values.
- Very specific: line 42
Audit log: ALWAYS write one line to the audit log on exit — success (shipped), escalation (escalated), idempotent skip (idempotent-skip), …
- Very specific: line 56
If user picks "skip-this-once-with-reason" → require an explicit reason recorded in notes.md, then continue.
- Very specific: line 58
If architecture.md exists, the new page must already be planned in it OR the user must add the row to the architecture before generating.
- Very specific: line 62
Read site-info.md → Geographic footprint.
- Distinctive: line 68
If invalid, ask user before proceeding.
- Distinctive: line 72
If user supplied → use it.
- Very specific: line 75
Architecture-fit check: If architecture.md exists, confirm this keyword has a planned row in it.
- Very specific: line 95
Before generating prose, produce a wireframe doc at sites/[site]/_drafts/[slug]/wireframe.md.
- Very specific: line 103
Catalog-footprint addition: if the site is catalog, also map the page's parent + sibling links per the architecture row (so …
- Very specific: line 124
Read site-info.md Languages section.
- Distinctive: line 128
Default coverage = every language declared in the Languages table.
- Distinctive: line 180
AI-search-friendly: self-contained facts: Every paragraph must stand on its own.
- Very specific: line 188
Prefer real work photos over stock when source is client-supplied or site-library.
- Distinctive: line 208
After shipping, check site's total service-page count vs.
“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
/service — Service page generator
Creates a production-ready service page for any service-business site managed in sites/.
Refusal up front
If sites/[site]/site-info.md → Business → Service business: false, the skill refuses to run and points the user to: "Use /blog for informational content, or generate a custom landing page outside the toolkit. Service pages are scoped to service businesses only."
Inputs
Required (asked at start if not given):
- Site — must match a folder under
sites/[name]/and haveservice-business: true. - Service — the service to feature (e.g., "Emergency Plumbing", "Dental Implants", "AI Consulting").
- Location (conditional on footprint):
single-location→ not asked; uses the one location insite-info.md.multi-location→ asked; must match a location insite-info.md.service-area→ asked; must be a city the business actually serves (persite-info.md).national-online→ not asked; no city.
Optional (per-run overrides):
- Primary keyword — defaults to picked-from-
service-keywords.csvmatching[service] [city]. - Image source — overrides default.
- Conversion template — overrides default if
notes.mddocuments a conversion winner.
Workflow
Step 0 — Cron-mode detection (if invoked with --cron)
If the invocation contains --cron, this skill runs in cron mode (no user available). Required behavior:
- Idempotency check: see this skill's "Routine versioning + idempotency contract" section. If today's output already exists, exit cleanly with
exit: "idempotent-skip". Write one line to the audit log viascripts/lib/audit-log.mjsappendRun({ exit: "idempotent-skip", ... }). - Escalation contract: any decision that would normally prompt the user (missing required file, voice anti-AI failure, schema validation failure, keyword cannibalization, etc.) MUST be escalated by writing an item to
sites/{site}/_inbox/viascripts/lib/cron-mode.mjswriteInboxItem(...). After writing, exit cleanly withexit: "escalated". Do NOT useAskUserQuestionin cron mode. - Defaults: when a choice would normally be asked, default to
site-info.md/goals.mdvalues. If both are silent on the required choice, escalate per (2). - Audit log: ALWAYS write one line to the audit log on exit — success (
shipped), escalation (escalated), idempotent skip (idempotent-skip), or failure (failed). - Backoff: at the start of every cron-mode run, call
checkBackoff({ routine })fromscripts/lib/audit-log.mjs. If true, the routine has hit the 3-strike threshold — write_inbox/routine-disabled-{name}.md, runscripts/sync-schedules.mjs --pause-routine={name}, and exit.
Step 1 — Read context (Rule 1)
Same as /blog Step 1, plus sites/[site]/service-keywords.csv and (if present) sites/[site]/architecture.md.
Step 1.5 — Architecture check (SEO_GUIDE.md Section 2.6)
Count rows in service-keywords.csv. If > 3 commercial keywords AND architecture.md is missing:
- Refuse to ship a new service page until
architecture.mdexists. - Output: "This site has [N] commercial keywords but no
architecture.md. Service pages need a planned hierarchy at this volume to avoid sprawl. Usetemplates/architecture.mdas the starting point. Would you like me to draft the architecture fromservice-keywords.csvnow? (yes/no/skip-this-once-with-reason)" - If user picks "skip-this-once-with-reason" → require an explicit reason recorded in
notes.md, then continue. The next/servicerun on this site re-prompts.
If architecture.md exists, the new page must already be planned in it OR the user must add the row to the architecture before generating. Off-architecture pages are how sites accumulate cannibalizing service pages.
Step 2 — Confirm footprint + scope
- Read
site-info.md→ Geographic footprint. - Confirm service + location combination is valid:
single-location→ service fromsite-info.mdservices list.multi-location→ service available at that location.service-area→ the city is in the service area list AND the service is offered.national-online→ service fromsite-info.mdservices list, no city.
- If invalid, ask user before proceeding.
Step 3 — Pick the keyword
- If user supplied → use it.
- Otherwise: pick the highest-CPC commercial keyword from
service-keywords.csvmatching the service + location, excluding any inused-keywords.md. - Seasonality check (soft) —
SEO_GUIDE.mdSection 2.5. If the picked keyword haspeak_monthsand/orseasonalitypopulated and today is out of the peak window for aseasonal/holiday-spiketerm, ORseasonality = declining, flag and ask before proceeding (same prompt format as/blogStep 2). Service pages in particular benefit from being live ~3 months before peak demand to give Google time to index and rank. Soft only — never refuse on seasonality alone. - Architecture-fit check. If
architecture.mdexists, confirm this keyword has a planned row in it. If not, ask the user to add it to the architecture before generating (a one-line append, then continue).
Step 4 — Determine URL pattern (Section 4.1 of SEO_GUIDE.md)
| Footprint | URL pattern |
single-location | /services/[service-slug] |
multi-location | /locations/[city-slug]/[service-slug] (or /services/[service-slug]/[city-slug]) |
service-area | /services/[service-slug]-[city-slug] |
national-online | /services/[service-slug] |
catalog | Hierarchical per architecture: /[category]/[subcategory]/[product-or-page]/ (Section 4.1.1). Path matches the architecture row's location in the tree. |
For catalog footprint sites, the URL is determined by where the keyword sits in architecture.md — Level 1 → /[category]/, Level 2 → /[category]/[subcategory]/, Level 3 → /[category]/[subcategory]/[leaf]/. If architecture.md shows the row uses a distribution page (Section 2.6.4), insert the axis marker: /[category]/[axis]/[value]/.
Step 5 — SERP analysis
Same workflow as /blog Step 4, but on the commercial keyword. Extract: word count (typically 1500+ for primary service pages, 800+ for city variants), H2/H3 outline, conversion elements (CTAs, trust signals, pricing, testimonials), schema patterns.
Step 5.5 — Wireframe (mandatory, per CLAUDE.md "Wireframe before content" rule)
Before generating prose, produce a wireframe doc at sites/[site]/_drafts/[slug]/wireframe.md. Mandatory contents:
- Layout zones: hero (H1 + value prop), trust signals strip, problem/pain section, solution sections (one per H2), social proof, FAQ, CTA + form, NAP block (per footprint).
- Heading map (H1 / H2 / H3) — H1 = primary commercial keyword; H2s = sub-aspects (what's included, who it's for, pricing, FAQ); H3s = items inside each H2.
- Internal linking pattern slots — for a service / leaf page, patterns 1 (header dropdown via template), 3 (footer via template), 4 (breadcrumb), 6 (related-services). Plus optional pattern 7 inbound from blog posts.
- Conversion element placement — above-fold CTA, sticky mobile CTA, click-to-call phone, trust signals position, multiple CTA placements.
- Image plan — hero + NAP + work photos (real client work for E-E-A-T) + city-specific hero for
service-areazipper pages. - NAP block placement — header / footer / contact section per
SEO_GUIDE.mdSection 4.3 (Local NAP rules by footprint). - Catalog-footprint addition: if the site is
catalog, also map the page's parent + sibling links per the architecture row (so internal-linking patterns 5 and 6 land cleanly).
Optional (if user opts in): hand-sketch / draw.io → AI design mockup for client presentation.
Show the wireframe to the user. Wait for explicit approval. Approved wireframes get archived in _drafts/[slug]/; rejected ones get revised and re-shown.
Step 6 — Plan + approval (Rule 2)
Present:
- Footprint + URL pattern + slug
- Primary keyword + cluster
- Required NAP for this page (per footprint, from
site-info.md) - H1, H2 outline, FAQ
- Conversion elements list
- Image source for this run
- Language fan-out (multilingual sites only) — list of languages this service page will be produced in (default: all declared in
site-info.md).
Wait for approval.
Step 6.5 — Resolve language fan-out (multilingual sites only)
Read site-info.md Languages section.
Multilingual: false→ skip; continue to Step 7 as a single-language run.Multilingual: true:- Default coverage = every language declared in the Languages table. Service pages almost always cover all languages — opt-out is rare and requires explicit confirmation.
- For each declared language:
- Resolve voice files: root
references/voice.md+ per-sitereferences/voice.[lang].md(REQUIRED). Service pages bias toward conversion-driven copy, lighter on stories — but still resolve all files. - Format URL using the language's URL pattern from the Languages table (substitute
[slug]). - Reserve a draft folder:
sites/[site]/_drafts/[slug]/[lang]/.
- Resolve voice files: root
- Generate the hreflang link cluster from the resolved language set (self-reference + all siblings +
x-default). - Schema
inLanguageis set per-language draft.Serviceschema getsinLanguagematching<html lang>.Service.availableLanguagemay list ALL languages the actual service is delivered in (a separate field — language of the page vs. languages the service is offered in). - Per-language nav coordination. The Lovable / CMS prompt for each language must include the new service page in that language's header dropdown / footer / sitemap. The English nav lists English service pages; the Spanish nav lists Spanish service pages.
The skill produces N drafts (one per language), each in its own _drafts/[slug]/[lang]/ subfolder. Each passes its own Tier 1 + conversion + voice anti-AI checks using its language's voice files.
Step 7 — Generate the page
Apply in order:
- Voice — resolved
references/, but biased toward conversion-driven copy (less story, more credibility). - Tier 1 on-page — all 16 items.
Anti-doorway-page rules (Section 4.1 of SEO_GUIDE.md) — for
service-areaonly:- Unique 200+ word opening specific to that city
- Local landmarks/neighborhoods mentioned
- 1–2 city-specific FAQs
- Real local NAP (city-specific area code if possible)
- Real testimonials from customers in that city
- Different hero image (not the same stock photo across cities)
If the page can't pass all 6, refuse to publish.
- NAP per footprint (Section 4.3 of SEO_GUIDE.md) — render correctly:
single-location→ site's one NAP, header + footer + contact section.multi-location→ that branch's NAP, with embedded map of that branch.service-area→ HQ address may be hidden; show service-area map; phone is local.national-online→ noLocalBusiness; onlyOrganizationcontactPoint.
- Conversion elements (Section 4.4 of SEO_GUIDE.md +
on-page-seo.mdCategory 14) — every item:- Above-fold CTA
- Sticky mobile CTA
- Phone with
tel:link - Trust signals above fold (rating, license, years)
- Multiple CTA placements
- Specific testimonials (names, photos, neighborhoods if
service-area) - Pricing transparency
- Service area / hours
- 4–8 FAQ
- Embedded map (per footprint)
- Schema (JSON-LD) —
Serviceschema (with all required properties fromon-page-seo.md9.3) +LocalBusiness(per footprint, from 9.2) +BreadcrumbList+FAQPage+Organizationsite-wide. Plus, IFF the site'ssite-info.mdbyline policy enables person-author mode AND the page'sauthor:frontmatter is set to a real-person slug (per/blogskill Step 6 item 9 routing logic), includePersonschema in the@graphand inject the bio block fromsites/{site}/author-{slug}.mdin the service page's author footer. Service pages traditionally don't carry author bylines, but with the partial-EEAT-byline policies (e.g., a site that adopted partial-EEAT-byline), service pages MAY carry a named author IF the service category is in the site'scoi-categories.md→personal_eligiblelist. Refer to/blogskill for the full author-routing logic — it's shared between the two skills. - Length — primary service pages 1500+ words; city variants 800+ words minimum.
- AI-search-friendly: Q+A density. Beyond the dedicated FAQ section (conversion element item 9), structure 3+ in-body sections as explicit question→answer pairs (the H2 or H3 IS a question; the paragraph below IS the answer in the first sentence). LLMs preferentially cite passages that stand alone as direct answers. Examples for service pages:
## How much does AI consulting cost for a 20-person team?(not## Pricing);### Do you work with companies outside Houston?(not### Geographic coverage). - AI-search-friendly: citation-friendly chunking. Paragraphs max 3 sentences. Lead each paragraph with the topic sentence — the main claim or value prop. Front-load named entities (the service name, location, key benefit) in the first sentence rather than burying them in subordinate clauses. Service pages have higher commercial intent and benefit even more from LLM-citable chunks.
- AI-search-friendly: self-contained facts. Every paragraph must stand on its own. NO "as mentioned above," "as discussed earlier," "see the section above," "we'll explain below." LLMs lose context between paragraphs — pretend each paragraph is the only one cited.
- AI-search-friendly: verifiable claims. Every statistic, customer count, response-time claim, or numeric assertion either (a) cites a source link, OR (b) is internally verifiable ("23 clients served in 2024" — provable via Stripe/GBP review counts/internal records, not McKinsey-style). Bare unsourced numbers read as fabricated. If you can't source or self-verify, drop the number.
Step 8 — Fetch images
Same as /blog Step 7, but with stricter rules:
- Hero image must be unique per city for
service-areazipper pages (can't reuse stock photo across all 50 city pages). - Prefer real work photos over stock when source is
client-suppliedorsite-library.
Step 9 — Validate (Rule 4)
Same as /blog Step 8, plus:
- ✅ Anti-doorway-page rules (all 6, for
service-areapages) - ✅ NAP consistency check (page NAP matches
site-info.mdexactly) - ✅ All required
Serviceschema properties present - ✅ Conversion elements all present (every checkbox in
on-page-seo.mdCategory 14 applicable)
Step 10 — Ship per publishing method
Same as /blog Step 9.
Step 11 — Update tracker
Append to used-keywords.md with Service page type.
Step 12 — Volume check (Section 4.2 of SEO_GUIDE.md)
After shipping, check site's total service-page count vs. cap per footprint. Flag if approaching the cap; recommend consolidation if multiple thin pages exist.
Routine versioning + idempotency contract
This skill participates in cruise-control via --cron mode. When invoked by cron:
- Stamp
routine_versionin every output produced (thenotes.mdaudit entry header,_inbox/item frontmatter, draft folder metadata, audit log line written viascripts/lib/audit-log.mjs). Currentroutine_version: 1.0. Bump when the skill's behavior meaningfully changes. - Idempotency: this skill MUST be safe to run twice in a row on the same day without producing duplicate work. Implementation: check
_drafts/{YYYY-MM-DD}-*at the start of every cron-mode run; if today's slug already exists, exit cleanly withexit: "idempotent-skip".
Reference: docs/specs/2026-05-16-agents-cruise-control-design.md § Operational hardening O3.
Refusal conditions
In addition to base refusals (missing files, etc.):
service-business: falseinsite-info.md.- Service or location not in
site-info.md. - Anti-doorway-page rules can't be satisfied (for
service-area). - Volume cap (Section 4.2) would be exceeded.
service-keywords.csvhas > 3 rows ANDarchitecture.mdis missing (SEO_GUIDE.mdSection 2.6) — refuse until the architecture exists, OR user records an explicit one-time skip reason innotes.md.
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.

