Skip to content

gooseworks-ai/goose-skills / drafts/marketing/learn-my-brand

learn-my-brand

Verified

Learn my brand once and stop asking me every time. Use when someone says learn my brand, set up my brand, here's our brand book, remember how we talk, what can we claim, can we say this, approve our claims, sign off our voice, our price or product changed, the client approved the claims, pick up where we left off, or check our claims are still current. Reads what GooseWorks already found about the brand, asks only the gaps as drafts to confirm, audits every claim (including ones research marked approved) and gets the ones that matter signed off with proof, scope and review date, turns the voice into rules the owner confirms, checks logo, colours and fonts, and saves it all as it goes so every other job loads it instead of asking. Ends with an unknowns list and a read-back. Not for who our customers are and what they say (use research-audience) or writing positioning from scratch (not covered yet).

references/category-rules.md

View on GitHub
# Category and platform rule packs

Load at step 3 when the brand's category is regulated, and at step 4 to seed the never-say list. Detect the category from the kit's `brandType`, the products and the owner's answer; confirm it with the owner in one line. Each pack lists default never-say entries, required qualifiers, and the claim types that default to "needs proof". Tag every rule you write with its source: **law**, **platform** or **owner**.

These are the common rules a senior marketer knows, not a legal opinion. Say so once. Never tell the owner a claim is "legal"; say it fits the usual rules or that it needs their legal review.

When a brand sells in several markets, record rules per market (US, UK, EU, India, etc.) and tag claims with the markets they are cleared for.

## Beauty and skincare (cosmetics)
- **Law.** The US line between cosmetic and drug is intended use. "Treats", "cures", "heals", "clears acne", "prevents breakouts", "anti-inflammatory", "boosts collagen", "repairs the skin barrier" make it a drug claim. Acne-treatment claims need an OTC-monograph active (salicylic acid, benzoyl peroxide, sulfur, adapalene) and drug labelling. EU: cosmetic claims follow the six common criteria (Reg. 655/2013). "Free from X" is barred when X is legal and normally present, or the claim denigrates it.
- **Default never-say:** treat, cure, heal, prevent (disease), anti-inflammatory, prescription-strength, medical-grade, "FDA-approved" (cosmetics are not approved), non-toxic, chemical-free, "safe for all skin", absolute "100% free of…" claims.
- **Safe alternatives:** "helps reduce the look of…", "visibly", "for acne-prone skin", "formulated without…" (with the list).
- **Needs proof by default:** "dermatologist-tested/-recommended/-developed", "clinically proven", "non-comedogenic", "hypoallergenic", percentage results.
- **OTC acne products are different.** A product with an acne-monograph active (salicylic acid 0.5–2%, benzoyl peroxide, sulfur, resorcinol or resorcinol monoacetate with sulfur) sold with a Drug Facts panel may lawfully say it treats acne, within the monograph's wording. Adapalene 0.1% is OTC through an approved drug application (an Rx-to-OTC switch), so its approved labelling sets what may be said. Ask once per product with such an active: "Is this sold as an OTC drug with a Drug Facts label?" Scope the never-say list per product: drug claims are banned for the cosmetics and allowed (in monograph wording) for the OTC drugs. A non-monograph active (e.g. mevalonic acid) claiming to treat acne is a violation either way.
- **Platform:** Meta bans copy that asserts or implies the viewer's personal attributes, including health conditions ("Struggling with acne?", "Your breakouts…"). Before/after imagery is restricted on Meta and TikTok.

## Supplements and wellness
- **Law.** US: structure/function claims only ("supports healthy sleep"), never disease claims ("treats insomnia"); the FDA disclaimer must accompany them ("These statements have not been evaluated by the Food and Drug Administration. This product is not intended to diagnose, treat, cure, or prevent any disease."). UK/EU: only authorised health claims from the register, in their authorised wording.
- **Default never-say:** cure, treat, prevent (disease), detox (as a medical effect), "boosts immunity" (EU unless authorised), weight-loss amounts, "doctor-recommended" without a survey.
- **Platform:** Meta and Google restrict weight-loss and health claims; no before/after for weight loss; personal-attributes rule as above.

## Food and beverage
- **Law.** Nutrient-content claims ("high protein", "low sugar", "good source of") have regulated thresholds. "Organic" needs certification. Allergen statements must be exact. Health claims follow the same rules as supplements.
- **Default never-say:** "healthy" (regulated in the US), "guilt-free", disease claims, "organic" without certification, "natural" without a definition the owner can stand behind.

## Alcohol, tobacco, cannabis/CBD
- **Law and platform:** age gating, no appeal to minors, no health or performance benefits, no suggestion of social or sexual success. CBD: no health claims; many ad platforms prohibit or restrict it by region.
- **Default:** every claim beyond product facts needs the owner's legal review.

## Finance (fintech, lending, investing, insurance, crypto)
- **Law.** US Reg Z: naming certain credit terms (a payment amount, a down payment, the number of payments) triggers required disclosures (APR and terms). UK FCA financial promotion rules: fair, clear, not misleading; risk warnings. US SEC Marketing Rule governs testimonials and endorsements for investment advisers.
- **Default never-say:** guaranteed returns, risk-free, "you will earn", "get rich", unqualified "no fees".
- **Platform:** Meta's special ad category for credit, employment and housing limits targeting; Google requires financial services verification in many regions.

## Health care providers and clinics
- **Law.** Patient testimonials and before/after need consent and may be restricted by professional bodies; no guarantees of outcome; HIPAA (US) for anything identifying a patient.
- **Default never-say:** guaranteed results, "painless", "best doctor", outcome promises.

## Kids and family
- **Law.** US COPPA for data from under-13s; UK CAP rules on advertising to children; toy safety claims need certification.
- **Default:** no direct calls to action aimed at children; no pressure to pester parents.

## B2B software and services
- **Law and practice.** Security and compliance claims (SOC 2, ISO 27001, HIPAA, GDPR-compliant) need the current certificate or report and its scope. Customer logos, names and case studies need written permission per customer; default none. Analyst quotes (Gartner, Forrester) have their own usage rules. Comparative claims against named competitors need like-for-like evidence. Roadmap items are internal until announced.
- **AI products.** Claims about what the AI does (accuracy rates, "never misses", "replaces your note-taker") need testing behind them; US FTC treats exaggerated AI capability claims as deceptive. Data claims ("we never train on your data", "encrypted", "GDPR compliant", "you own your data") must match the actual terms and data-processing agreement.
- **Recording and notes tools.** Call-recording consent rules (one-party vs all-party consent states and countries) shape what may be promised ("record any call").
- **Design partners and beta customers** are a material connection: their quotes need disclosure where relevant, and each name needs written permission (who granted it, when, how). Anonymised descriptions ("a 200-person fintech") must be accurate and permitted too.
- **Default never-say:** "#1" without a source, "enterprise-grade security" without the report, "SOC 2 certified", uptime numbers above the SLA, unannounced features, unpermissioned customer names.

## Professional services (legal, accounting, financial advice, medical)
- **Law.** Many professional bodies restrict testimonials, superlatives ("best lawyer") and outcome guarantees. Financial advisers fall under the SEC Marketing Rule (US) or FCA rules (UK).
- **Default never-say:** guaranteed outcomes, "specialist" or "expert" where regulated titles apply, client names without consent.

## Consumer apps
- **Platform.** App Store and Google Play guidelines bar misleading metadata, fake ratings, and references to other platforms in screenshots. Store ratings are per platform and per country: tag them. Free trial terms (length, price after, cancellation) must be clear.
- **Law.** Kids: COPPA. Health, wellness and habit data: US FTC Health Breach Notification Rule and state laws such as Washington's My Health My Data Act; GDPR special-category data (EU). Privacy promises must match the privacy policy.
- **Pre-launch:** beta-tester quotes need permission and carry a material connection (free access); the launch date defaults to embargoed until the founder says otherwise; "science-backed" or "proven" habit claims need the science named.

## Marketplace sellers (Amazon, Etsy, Walmart, Flipkart)
- **Platform.** Amazon bars claims about competitors, time-sensitive language ("sale", "best seller" unless earned), and health claims for many categories; its rules are stricter than the brand's own site. Scope each claim to the channel. Record authorised sellers and any minimum advertised price policy.
- The agent cannot read Amazon pages automatically. Ask the owner to paste or upload listing copy.

## Environmental claims (any category)
- US FTC Green Guides; UK CMA Green Claims Code; EU Directive 2024/825, in application from 27 September 2026: generic claims ("eco-friendly", "green", "climate-friendly") without recognised excellent performance are banned, and product claims of carbon neutrality based on offsets are banned.
- **Default never-say:** eco-friendly, sustainable, green, carbon neutral, "good for the planet", unless specific, certified and scoped.

## Advertising platforms (any category)
- **Meta:** no personal-attributes copy (health, financial status, religion, sexual orientation, etc.); restricted categories (health, weight loss, alcohol, dating, finance, politics); ads must match the landing page.
- **Google:** misrepresentation policy (unclear pricing, unsubstantiated claims); healthcare and medicine restrictions; trademark use in ad text can be limited by the owner's complaint.
- **TikTok:** stricter on before/after, weight loss, and health claims; branded-content disclosure.
- **Creators and influencers (any platform):** disclosure of material connections (#ad, paid partnership) is required by US FTC, UK CMA/ASA and most EU regulators.
rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first.Stars: 194kUpdated Claude CodeCodexVerifiedInstall