WORKPULSION: TWO SMALL EDITORIAL TRIALS
https://workpulsion.com/guides/claude-code-skill-repositories/

FICTIONAL NORTHLINE HOMEPAGE DRAFT
Editorial trial, October 12, 2026. One Codex session manually followed the
copywriting procedure from marketingskills at 1efedbc5148b54b2f0f6c6c9fe0be62e151c7fff.
No collection/plugin installation, automatic-invocation test or conversion measurement.

PRIMARY HERO
Headline: A clear website for your service business
Subheadline: Present your services, collect enquiries and link visitors to your existing booking service.
CTA: Request a scope review

SUPPORTING SECTION
Heading: Agree the pages before the build
Body: Start with Home, Services and Contact. We discuss what each page needs and where enquiries should go.
CTA: Request a scope review

HEADLINE AND CTA OPTIONS
A. A clear website for your service business / Request a scope review
Rationale: names the audience and the initial action.
B. Give customers a place to explore your services / Discuss your website scope
Rationale: describes the visitor's task without inventing performance evidence.
C. Put your services and enquiry details in one place / Plan your website pages
Rationale: stays within the approved three-page offer.

ANNOTATIONS
The primary CTA describes a scope discussion rather than promising a finished quote.
The booking service is an existing external service; custom booking is excluded by the brief.
Social proof is omitted from the public draft: [NEED: an approved customer example or testimonial].
No numerical performance claim, delivery deadline or free-trial promise is supported by the brief.
Meta content is deferred until the real business identity and service location are confirmed.

SELF-REVIEW
Reviewed the hero, section and alternatives for the procedure's banned patterns.
No contrast reveal, negation list, rhetorical reveal, em dash or trailing pile-on in draft copy.
Draft sentences are at most 17 words. No exclamation marks or unsupported figures.
Audience and booking detail tie the copy to the brief; the broad headline still has limited distinctiveness.
[NEED: a factual differentiator before final commercial use].
Text cadence was reviewed silently; no audio/read-aloud review was performed.
This is a reviewable first draft, not evidence of better conversions or superiority over a plain request.


WEBSITE VERIFICATION
A manually applied verification procedure was used to identify relevant checks, run our checker and inspect its fresh output. The checker is Workpulsion material, not supplied by superpowers.

{
  "date": "2026-10-12",
  "execution": "instructions manually applied in current Codex session; no plugin installation or automatic trigger test",
  "python": "3.14.4",
  "marketing_checks": {
    "approved_primary_cta": true,
    "existing_booking": true,
    "three_options": true,
    "no_unsupported_performance": true
  },
  "website_checks": [
    {
      "file": "before.html",
      "checks": [
        "approved canonical",
        "local link targets"
      ],
      "errors": [
        "Canonical points outside approved fictional origin",
        "Missing local link target: missing.html"
      ],
      "result": "FAIL",
      "exit_code": 1
    },
    {
      "file": "after.html",
      "checks": [
        "approved canonical",
        "local link targets"
      ],
      "errors": [],
      "result": "PASS",
      "exit_code": 0
    }
  ],
  "limits": "No baseline comparison, Claude Code runtime test, user study or full website audit."
}


REPRODUCTION AND SCOPE
WORKPULSION SYNTHETIC EDITORIAL TRIALS
October 12, 2026

The two procedures were manually applied in a Codex session. No plugin installation or Claude Code runtime test. No baseline comparison or conversion measurement.

Marketing input: marketing/.agents/product-marketing.md
Marketing output: marketing/output.txt
Website input and correction: review/before.html and review/after.html
Our checker: review/check.py (not a third-party script)
Recorded results: results.json

To repeat the two website checks with Python, from the review folder run:
python check.py before.html
Expected: exit 1, wrong canonical and missing Contact target.
python check.py after.html
Expected: exit 0, both limited checks pass.

From the bundle root, python run.py repeats both checks and checks four aspects of the supplied marketing draft. It does not generate new copy or invoke any skill.

Exact source revisions reviewed:
{
  "coreyhaines31/marketingskills": "1efedbc5148b54b2f0f6c6c9fe0be62e151c7fff",
  "obra/superpowers": "bb92a77741419a4ab5f06e711a283343f1ada0c3",
  "anthropics/skills": "dbd4588f9e1033efb41dad4bef2f7947c8993d44",
  "AgriciDaniel/claude-seo": "4b99de2f7de7e7d5247042e5fb5b4ea9368ef734"
}

Third-party sources are linked in the article, not redistributed in this bundle.