Sota copywriting
State-of-the-art marketing copywriting and content guidance (2026) covering positioning and value propositions, headlines and landing-page copy, CTAs and social proof, SEO content (search intent, E-E-A-T, Google spam policies), and the accuracy/legal layer — claim substantiation, FTC Endorsement Guides (2023) and the Consumer Reviews Rule (2024), testimonial consent, dark-pattern avoidance, and email-marketing law (CAN-SPAM, GDPR consent). Use when writing or reviewing ANY outward-facing content — landing pages, marketing sites, product pages, launch announcements, app-store listings, newsletters, cold/lifecycle email — AND when auditing existing copy for conversion quality, honesty, and legal exposure. Trigger keywords: copywriting, marketing copy, landing page, headline, CTA, value proposition, positioning, tagline, SEO, meta description, E-E-A-T, testimonial, social proof, case study, newsletter, email marketing, app store listing, launch post, pricing page, brand voice.From its SKILL.md
npx -y skills add martinholovsky/SOTA-skills --skill sota-copywritingAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 12 stars12 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
7.4 KB, ~1.5k tokens by cl100k_base, as published. Nobody here has run it
SOTA Copywriting & Marketing Content
Purpose
Expert-level rules for the words that face the market: landing pages, product marketing, SEO content, launch posts, and lifecycle email. The core thesis: conversion copy is a claim-making machine, and every claim is both a persuasion asset and a legal/trust liability — the craft is being specific enough to convince while staying provable.
Boundaries: this skill owns outward-facing content. In-product interface text
is sota-ux-writing; technical documentation is sota-docs-workflow; the
visual/layout layer of marketing pages is sota-frontend-design; consent and
privacy-notice engineering is sota-privacy-compliance.
BUILD mode
When writing marketing/content copy:
- Define audience, awareness stage, and the one action before drafting —
a page without a single primary conversion goal is a brochure (
rules/01). - Lead with the outcome the reader buys, prove it with features and
evidence — benefit → mechanism → proof, in that order (
rules/01§3). - Draft headlines last, pick from ≥5 variants, test against the 4 U's
heuristic, and keep the above-fold contract: what it is, who it's for,
what to do next (
rules/02). - Run the accuracy gate before publishing (
rules/04§1): every objective claim substantiated, every superlative provable or cut, every testimonial consented and typical, every disclosure placed where it's seen. - For anything meant to rank: match search intent and write people-first
— E-E-A-T signals on the page, no scaled low-value content (
rules/03). - Before finishing, run each loaded rules file's Audit checklist against the draft.
AUDIT mode
When reviewing existing marketing content:
- Inventory the surface: landing pages, pricing page, comparison pages, blog/SEO content, email templates/sequences, app-store listings, social proof assets.
- Claims-first pass (
rules/04): list every objective claim, superlative, statistic, testimonial, and urgency device; demand the substantiation for each. Legal exposure outranks conversion polish. - Conversion pass (
rules/01–02): above-fold contract, message hierarchy, CTA quality, proof placement, scannability. - SEO pass (
rules/03) where organic traffic matters: intent match, E-E-A-T, spam-policy exposure, title/meta/heading hygiene.
Severity conventions
- Critical — legal/deception exposure: unsubstantiated objective claims, fake or undisclosed-material-connection testimonials, fabricated reviews or social-proof counts, fake urgency (countdown timers that reset), missing unsubscribe/consent basis in email, dark patterns.
- High — trust or ranking damage: superlatives without proof, "results not typical" testimonials presented as typical, spam-policy-violating content (scaled/doorway), misleading comparison tables, meta/title bearing no relation to page content.
- Medium — conversion damage: no clear above-fold value proposition, feature-first copy with no benefit framing, generic CTAs ("Learn more"), proof present but unspecific, intent-mismatched SEO content.
- Low — polish: weak headline variants, wall-of-text sections, missing message match between ad and landing page, stale dates.
Finding format
file:line | rule violated | severity | effort | fix — for claim findings,
state what substantiation would be sufficient; for copy findings, propose the
rewritten line.
Rules index
| File | Read this when... |
|---|---|
rules/01-positioning-value-proposition.md | Deciding what the copy argues: audience and awareness stages, value-proposition structure, benefit-vs-feature discipline, message hierarchy, differentiation claims, evidence types, we-vs-you framing. |
rules/02-headlines-landing-pages.md | Writing the page: headline craft and variants, above-fold contract, section flow (attention→interest→desire→action as a heuristic), scannability, CTA wording and placement, social-proof presentation, pricing-page honesty. |
rules/03-seo-content.md | Content meant to be found: search-intent matching, people-first content and E-E-A-T, Google spam policies (scaled content abuse, site reputation abuse, doorway pages), titles/meta as display heuristics, structured-data honesty, AI-assisted content discipline, generative-AI search surfaces (AI Overviews/AI Mode — still SEO, no GEO folklore), content maintenance. |
rules/04-claims-legal-trust.md | The accuracy/legal gate: claim substantiation, puffery vs objective claims, FTC Endorsement Guides (16 CFR 255) and Consumer Reviews Rule (16 CFR 465), testimonial consent and typicality, dark patterns, email law (CAN-SPAM, GDPR/ePrivacy), security/uptime claim hygiene. |
Top-10 non-negotiables
- Every objective claim is substantiated before it ships — numbers have sources, "only/first/fastest" has evidence, or the claim is cut. (rules/04 §1)
- No fake anything: reviews, testimonials, follower counts, urgency timers, "X people are viewing this". Fabricated social proof is a Critical finding and, for reviews, a civil-penalty violation (16 CFR 465). (rules/04 §3–4)
- Material connections are disclosed clearly and proximately — paid, gifted, employee, or affiliate relationships behind any endorsement. (rules/04 §2)
- Benefit first, feature as proof — every section passes the "so what?" test from the reader's chair. (rules/01 §3)
- One page, one primary action; every CTA is verb + outcome ("Start free trial"), never "Submit" or a bare "Learn more". (rules/02 §4)
- The above-fold contract holds: a first-time visitor can answer "what is it, who's it for, what do I do next" without scrolling. (rules/02 §2)
- Skim test passes: reading only headings tells the page's whole argument. (rules/02 §3)
- Search content matches intent and adds first-hand value — no scaled thin content, no doorway pages, no ranking-borrowed third-party content (Google spam policies). (rules/03 §1, §3)
- Email requires a basis and an exit: consent (or a lawful basis) to send, accurate subject/sender, and a working unsubscribe honored promptly. (rules/04 §5)
- Friction reducers must be true: "No credit card required", "Cancel anytime", "5-minute setup" are claims, not decorations — verify or delete. (rules/02 §4, rules/04 §1)
What ships with it: 4 files
34.5 KB alongside SKILL.md
rules/
Gives 0 of the 12 instructions most marketing audience skills give in ~1.5k tokens
Counted across 690 of the 894 authors here whose files we hold, read 2026-08-07
- Apply Poppins font to headingsin 41 of 690, across 6 files
- Apply Lora font to body textin 41 of 690, across 6 files
- Use Arial fallback for headingsin 39 of 690, across 4 files
- Use Georgia fallback for body textin 39 of 690, across 4 files
- Maintain text hierarchy and formattingin 39 of 690, across 4 files
- Use accent colors for non-text shapesin 38 of 690, across 3 files
- Use RGB values for precise color matchingin 38 of 690, across 3 files
- Use brand colors for primary text and backgroundsin 36 of 690, across 1 file
- Read product marketing context file before asking questions, starting, or auditingin 35 of 690, across 23 files
- Use active voice instead of passive voicein 26 of 690, across 10 files
- Implement or generate appropriate JSON-LD structured datain 24 of 690, across 17 files
- Prioritize clarity over clevernessin 22 of 690, across 8 files
Said here and by no other author read
- Define audience, awareness stage, and one action before drafting
- Lead with the outcome, then prove it with features
- Draft headlines last from five or more variants
- Run the accuracy gate before publishing
- Match search intent and write people-first content
- Run each loaded rules file audit checklist against the draft
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.