Gerador de Estratégia de Conteúdo Inicial
wolfganghuang0083/content-strategy-starter
De briefing básico para estratégia de conteúdo e mapa de 25 tópicos em uma execução.
Ver no GitHubSkill que transforma informações básicas de empresa/produto/serviço em estratégia de conteúdo completa: 5 personas, cenários JTBD, títulos com intenção de busca e mapa de conteúdo inicial com 25 tópicos. Permite que sites novos sem histórico de dados lancem marketing de conteúdo imediatamente.
Instalação
npx convez-ai wolfganghuang0083/content-strategy-starter1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Lançar conteúdo em sites novos sem dados históricos de busca
- Mapear personas e cenários JTBD para produtos B2B ou B2C
- Gerar títulos que interceptam intenção de busca real
- Criar seed content map para validar demanda antes de escalar
Veredito

Ferramenta prática que resolve o problema real de startups sem data histórica. Combina método de descoberta de demanda com escrita de headlines orientada a conversão, funcionando como framework estruturado e não apenas prompt genérico. Execução clara via CLI, com exemplos e integração em pipeline maior.
README
A tradução para o português deste README ainda está na fila.
content-strategy-starter
📖 手把手 fork + 使用教學(GitHub Pages):https://wolfganghuang0083.github.io/content-ops-toolkit/
The Step 0 of the content-ops toolkit. Give it basic company / product / service info and it produces a full content strategy from scratch — personas, demand scenarios, search-intent titles, and a seed content map — so a brand-new site with zero keywords and zero Search Console data can still launch content marketing.
Growth Hacker × Master Copywriter in one skill. GH — demand discovery: derive the keywords people actually type from
persona × job-to-be-done scenarios, sidestepping the "no historical data" chicken-and-egg. Copywriter — headline interception: turn "this person, in this moment, stuck on this problem" into "the title they click right now."
Where it sits
content-strategy-starter (basic info → personas/queries/titles) ← YOU ARE HERE
→ content-map-builder (rank into a data-driven map)
→ content-pipeline (write & draft) → content-schedule (when to publish) → internal-linking (weave the cluster)
What it produces (the 6-step framework)
- What kind of site is this — type, model, positioning, conversion goal.
- 5 buyer personas — distinct enough that their search queries don't overlap.
- JTBD demand matrix — per persona: scenario → problem → search trigger → the actual query → funnel stage.
- Feature → problem mapping — which feature solves which persona's problem (find your core vs weak selling points).
- Search-intent titles — headlines that catch the moment of search (copywriter rules built in).
- 25-topic bank (5 personas × 5 scenarios) → de-duped, clustered by strategic role → a seed content map ready for content-map-builder.
Run it (one shot)
Workflow({ scriptPath: '~/.claude/skills/content-strategy-starter/scripts/generate_strategy.js',
args: { business: '<paste your basic brief, or one paragraph about the company/product/service>',
market: '台灣繁體中文', // or 'US English'
personas: 5, scenariosEach: 5,
outPath: '/tmp/strategy.md' } })
Then read outPath → strategy doc + 25-topic bank + seed content map. Paste the seed map into content-map-builder and continue down the chain.
Or just tell Claude: "用 content-strategy-starter 幫 <這個生意> 規劃一批內容題目。"
Files
| File | Purpose |
|---|---|
SKILL.md | The 6-step method (GH + copywriter framing) + rules. |
scripts/generate_strategy.js | The automation — basic info → personas → scenarios → titles → seed map. |
templates/business-brief.template.md | The only input you fill in. |
examples/example-strategy.md | A worked example (fictional pet-food brand, 5 personas × 5 = 25 titles). |
License
MIT.