Growth content core
Shared reference for the growth-content cluster: the searchable-vs-shareable decision, the buyer-stage map every asset is aimed at, the one product-marketing-context source of truth all spokes read, voice-of-customer conventions, and the asset × stage routing matrix. USE WHEN planning a campaign, picking a content type, or writing any growth asset — the interlocking rules every spoke shares.From its SKILL.md
npx -y skills add Sheshiyer/skill-clusters --skill growth-content-coreAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 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
6.4 KB, ~1.4k tokens by cl100k_base, as published. Nobody here has run it
Growth Content Core
Shared model for the growth-content cluster. The strategy, writing, conversion, and launch
spokes all depend on these decisions — keep them consistent here so no spoke contradicts another.
1. The decision this domain turns on: searchable vs. shareable
Every asset must be searchable, shareable, or both — and you decide before you write. Prioritize in that order; search demand is the foundation that compounds, shareable demand is the spark that spreads.
captures EXISTING demand creates NEW demand
(people already looking) (spread an idea)
SEARCHABLE ──────────────────────────┐ ┌────────────────────── SHAREABLE
keyword-led · intent-matched │ │ insight-led · emotion-driven
comprehensive · structured ▼ ▼ novel · contrarian · story
┌──────────────────────────────┐
│ BOTH = the strongest asset │
│ (rank AND get passed around) │
└──────────────────────────────┘
- Searchable → target a specific query, match intent exactly, structure for skim + LLM citation. Owned by
content-strategy(planning) andcontent-research-writer(writing). - Shareable → lead with a novel insight, original data, or a take worth repeating. Same two spokes, different mode.
- If an asset is neither, it has no distribution. Send it back to
content-strategy.
Rule: name the call (searchable / shareable / both) at the top of every brief. It dictates the headline, the structure, and the channel.
2. The buyer stage every asset is aimed at
One asset serves one stage. The stage sets the keyword modifiers, the CTA, and the proof.
| Stage | Modifiers / intent | Asset that fits | Primary spoke |
|---|---|---|---|
| Awareness | "what is," "how to," "guide to" | educational blog, checklist, quiz | content-research-writer, lead-magnets |
| Consideration | "best," "vs," "alternatives," "comparison" | comparison page, assessment, webinar, ad | copywriting, ad-creative, lead-magnets |
| Decision | "pricing," "demo," "trial," "templates" | pricing copy, ROI tool, sales email | copywriting, free-tool-strategy, cold-email |
| Retention/Expansion | onboarding, upsell, win-back | lifecycle email, usage nudge | email-sequence |
Mismatched stage is the most common failure: a decision-stage CTA on an awareness asset converts no one. State the stage in the brief.
3. The one source of truth: product-marketing-context
product-marketing-context writes .agents/product-marketing-context.md — positioning, ICP,
JTBD, competitors, differentiation, objections, voice-of-customer language, proof points,
brand voice. Every spoke reads it first. It exists so the user never re-explains their product
and so all output stays consistent.
- New project → run
product-marketing-contextbefore any other spoke. - It exists → spokes pull voice + proof from it and only ask for task-specific gaps.
- Missing on an older setup → check
.claude/product-marketing-context.mdand offer to move it.
4. Voice & evidence conventions (shared by all spokes)
- Customer language > company language. Mirror verbatim phrases from reviews, calls, support tickets. Exact words beat polished descriptions.
- Specific > vague. "Cut reporting from 4 hours to 15 minutes," not "save time." Numbers, timeframes, names.
- Benefits > features. Every feature gets a "which means…" bridge to an outcome.
- One idea per unit. One CTA per page/email, one job per asset, one argument per section.
- Proof is sourced, never invented. Pull stats, testimonials, and logos from the context doc or ask for them. Fabrication is a hard stop.
- Banned filler: "leverage," "synergy," "best-in-class," "seamless," "cutting-edge," "utilize," "robust," exclamation points, and "I hope this email finds you well."
5. Asset × funnel routing matrix
| Funnel stage | Plan | Write | Convert | Launch |
|---|---|---|---|---|
| Spoke(s) | content-strategy, marketing-ideas, marketing-psychology | content-research-writer, copywriting, copy-editing | lead-magnets, free-tool-strategy, cold-email, email-sequence, ad-creative | launch-strategy |
| Output | pillars, calendar, ideas, principles | articles, page copy, edited copy | gated assets, tools, outbound + lifecycle email, ads | GTM plan, Product Hunt, rollout |
| Reads context doc | yes | yes | yes | yes |
Standard pipeline: strategy → write → edit → convert → launch. Cold outbound (cold-email)
and lifecycle (email-sequence) are distinct lanes — cold = no relationship, lifecycle = existing
contact. Downloadable (lead-magnets) vs. interactive (free-tool-strategy) is the other common fork.
6. Shared guardrails
- Searchable, shareable, or both — name it before writing; no asset ships without a distribution path.
- One asset, one stage, one job, one CTA.
- Customer language, never buzzwords; read it aloud — if it sounds like marketing, rewrite it.
- Proof is sourced, never invented — hard stop on fabricated stats/testimonials/logos.
- Draft, then
copy-editing— the seven sweeps run on a complete draft, not midstream. - Urgency/scarcity only when genuine.
- Context first — pull voice and proof from
product-marketing-context.md; if it's missing, that's the first task. - Sibling clusters:
business-content(voice-profile-led GTM writing — articles, investor materials, outbound),social-media(channel-native social). Reach for those when the work is brand-voice or platform-specific rather than demand-generation.
What ships with it
Read from the repository
Just SKILL.md. No reference files, no scripts.