据 AlphaMoat 统计,截至 2026-09,Lead Gen Website Builder (SEO PUR) 累计下载 3,832 次、获 12 个星标,在全部 5,078 个 AI 技能中排名第 2,735 位,在「开发编程」分类(577 个)中排名第 355 位。
| 平台 | 累计安装 | 累计下载 | 星标 | 评论 |
|---|---|---|---|---|
| ClawHub ↗ | 120 | 3,653 | 12 | 2 |
| SkillHub ↗ | 251 | 3,832 | 12 | — |
各平台的计数口径不同,统计时间不一致。
为本地服务型企业规划并构建获客网站,涵盖服务页面、SEO、结构化数据、转化追踪、分析和合规内容。上线前应核对业务事实、隐私与法律文本、追踪设置以及本地平台政策。
资料发布于 2026-10-27
以下是根据技能文档生成的示例,不代表实际运行结果。
2026-09 起统计方法调整:同名技能按作者分别统计,并覆盖全部技能。图中虚线为调整前的月份,与之后的数据不直接比较,也不计算跨越调整的增长。
| 月份 | 累计下载 | 环比 | 星标 |
|---|---|---|---|
| 2026-09 | 3,832 | — | 12 |
| 2026-08 | 3,574 | +4.7% | 12 |
| 2026-07 | 3,415 | +7.2% | 12 |
| 2026-06 | 3,187 | +4.1% | 12 |
| 2026-05 | 3,061 | — | 11 |
| 2026-03 | 2,720 | — | 10 |
以下是作者发布的原始文件,按纯文本显示,本站未做修改或审核;其中的链接不可点击,命令请核实后再运行。抓取于 2026-10-23。
--- name: lead-gen-website description: Build complete local lead generation websites with SEO optimization, conversion tracking, and RGPD compliance. Use for creating service-based websites targeting local markets (plumbers, electricians, home services, etc.) with 10-20 pages, structured data, analytics, and legal compliance. --- # Lead Generation Website Builder Build conversion-optimized local service websites with complete SEO, tracking, and RGPD compliance — **avec garde-fous anti-spam (Google Spam Policies + March 2024)**, local SEO (GBP) et micro-budget ads. ## When to Use This Skill
Use this skill when the user requests a website for:
- Local service businesses (home services, repairs, professional services)
- Lead generation focused on specific geographic areas
- Sites requiring 10-20+ pages with service pages, blog, and legal pages
- SEO-optimized content targeting local keywords
- Conversion tracking (phone, WhatsApp, forms with UTM parameters)
- RGPD/GDPR compliance (cookie banner, privacy policy, legal pages)
## Workflow Overview
Follow these phases sequentially. Do NOT skip phases or combine them without clear reason.
0) **Policy / Risk Check (mandatory)**
- Read `references/google-spam-guardrails-2024.md`
- Explicitly verify: no doorway pages, no scaled generic content, no fake address/avis, no misleading claims.
- If the project is **mise en relation** (leadgen): require clear disclosure on all key pages.
Then continue with Phases 1→7.
### Phase 1: Analysis and Planning
Gather project requirements from the user or specifications document.
**Required information:**
- Business niche and services offered
- Geographic target area (city + radius)
- Target keywords for SEO
- Contact information (phone, WhatsApp, email)
- Number and types of pages needed
- Competitor websites (for differentiation)
**Output:** Clear understanding of project scope, target audience, and conversion goals.
### Phase 2: Design Brainstorming
Create `ideas.md` in the project root with THREE distinct design approaches.
Use `templates/design-ideas-template.md` as structure. Each approach must define:
- Design movement and aesthetic philosophy
- Color palette with hex codes and emotional intent
- Typography system (headings + body fonts)
- Layout paradigm (avoid generic centered layouts)
- Signature visual elements
- Animation guidelines
- Interaction philosophy
Consult `references/design-philosophies.md` for inspiration, but create original combinations.
**Selection:** Choose ONE approach and document the rationale. This design philosophy will guide ALL subsequent design decisions.
### Phase 3: Visual Assets Generation
Generate 3-5 high-quality images using `generate` tool. These images MUST:
- Align with the chosen design philosophy (colors, mood, style)
- Be stored in `/home/ubuntu/webdev-static-assets/`
- Cover key visual needs: hero background, service illustrations, local landmarks, team/artisan photos
Plan strategic usage:
- Hero section: Most impactful image
- Service pages: Relevant illustrations
- About/Trust sections: Team or local landmark photos
Do NOT generate images on-the-fly during development. Generate all at once for efficiency.
### Phase 4: Content Structure
Create detailed content structure for all pages.
**Option A (Manual):** Write `content-structure.md` directly with sections for each page including title, meta description, H1, and main content outline.
**Option B (Script):** Create `specs.json` with page data, then run:
```bash
python /home/ubuntu/skills/lead-gen-website/scripts/generate_content_structure.py specs.json content-structure.md
```
**Content requirements:**
- Minimum 500 words per main page (homepage, main services)
- Minimum 1000 words per blog article
- Include target keywords naturally (no stuffing)
- Answer user intent (what, why, how, cost, area)
- Local focus (mention city/region frequently)
### Phase 5: Development
Initialize project and build all pages.
#### 5.1 Initialize Project
```bash
webdev_init_project <project-name>
```
#### 5.2 Configure Design Tokens
Edit `client/src/index.css` with chosen design philosophy:
- Update CSS variables for colors (primary, secondary, accent, background, foreground)
- Configure typography (font-family for sans, serif)
- Adjust shadows, radius, animations
Add Google Fonts in `client/index.html`:
```html
<link rel="preconnect" href="https://fonts.googleapis.com" />
<link rel="preconnect" href="https://fonts.gstatic.com" crossorigin />
<link href="https://fonts.googleapis.com/css2?family=YourFont:wght@400;600;700&display=swap" rel="stylesheet" />
```
#### 5.3 Create Reusable Components
Use templates from `templates/` directory. Replace placeholders with project-specific values:
**Header** (`templates/component-Header.tsx`):
- `{{SITE_NAME}}`, `{{SITE_TAGLINE}}`, `{{SITE_INITIALS}}`
- `{{PHONE_NUMBER}}`, `{{WHATSAPP_NUMBER}}`
- `{{NAV_ITEMS}}` (JSON array of `{label, href}`)
**Footer** (`templates/component-Footer.tsx`):
- `{{SITE_NAME}}`, `{{SITE_DESCRIPTION}}`
- `{{SERVICE_LINKS}}`, `{{UTILITY_LINKS}}`
- `{{PHONE_NUMBER}}`, `{{EMAIL}}`, `{{LOCATION}}`
**SEOHead** (`templates/component-SEOHead.tsx`):
- Replace `{{DOMAIN}}` with actual domain
**Other components:** Breadcrumbs, ContactForm, CookieBanner (copy as-is, minimal customization needed)
#### 5.4 Build Pages
**For similar pages (services, blog articles):**
1. Create template file (e.g., `service-template.tsx`) using `templates/page-service-template.tsx`
2. Create data file (e.g., `services-data.json`) with array of page data
3. Run batch generation:
```bash
python /home/ubuntu/skills/lead-gen-website/scripts/generate_pages_batch.py service-template.tsx services-data.json client/src/pages/
```
**For unique pages (homepage, tarifs, FAQ, contact):**
Build manually with rich, custom content. Use components for consistency.
**For legal pages:**
Use `templates/page-legal-template.tsx` with standard legal content.
#### 5.5 Update App.tsx
Add all routes to `client/src/App.tsx`:
```tsx
<Route path="/service-page" component={ServicePage} />
```
Integrate Header, Footer, and CookieBanner in App layout.
### Phase 6: SEO, Tracking, GBP, Ads
#### 6.1 Generate SEO Files
Create `pages.json` with all URLs and priorities:
```json
[
{"url": "/", "priority": "1.0"},
{"url": "/service", "priority": "0.9"},
{"url": "/contact", "priority": "0.9"},
{"url": "/blog", "priority": "0.6"},
{"url": "/mentions-legales", "priority": "0.3"}
]
```
Run script:
```bash
python /home/ubuntu/skills/lead-gen-website/scripts/create_seo_files.py yourdomain.com pages.json client/public/
```
This creates `robots.txt` and `sitemap.xml` in `client/public/`.
#### 6.2 Add Structured Data
Add JSON-LD structured data to key pages using SEOHead component's `jsonLd` prop:
**Homepage (LocalBusiness):**
```tsx
const jsonLd = {
"@context": "https://schema.org",
"@type": "LocalBusiness",
"name": "Business Name",
"telephone": "+33123456789",
"email": "[email protected]",
"address": {
"@type": "PostalAddress",
"addressLocality": "City",
"addressCountry": "FR"
},
"areaServed": ["City1", "City2"],
"openingHours": "Mo-Fr 08:00-18:00"
};
```
**Service pages (Service):**
```tsx
const jsonLd = {
"@context": "https://schema.org",
"@type": "Service",
"name": "Service Name",
"description": "Service description",
"provider": {
"@type": "LocalBusiness",
"name": "Business Name"
},
"areaServed": "City"
};
```
Consult `references/seo-checklist.md` for complete SEO requirements.
#### 6.3 RGPD Compliance
Verify:
- CookieBanner component is integrated in App.tsx
- Privacy policy page exists with complete RGPD information
- Cookie policy page exists
- Legal mentions page exists
- ContactForm includes link to privacy policy
Consult `references/rgpd-compliance.md` for detailed requirements.
#### 6.4 GBP / Local SEO (Prominence)
Read and apply:
- `references/gbp-local-seo-playbook.md`
Deliverables to produce:
- GBP setup checklist (catégories/services/Q&R)
- 30-day photo/post/avis plan
- NAP citations list (quality-first, no spam)
#### 6.5 Micro-budget Ads (4€/jour)
Read and apply:
- `references/ads-micro-budget-4eur-playbook.md`
Deliverables to produce:
- 1 campagne ultra-serrée (keywords exact/phrase, zone, horaires, négatifs)
- 1 landing dédiée + tracking
#### 6.6 Conversion Tracking
ContactForm component automatically captures UTM parameters from URL:
- `utm_source` (e.g., google, facebook)
- `utm_campaign` (campaign name)
- `utm_adset` (ad set name)
- `utm_ad` (specific ad)
These are stored in form state and can be sent to backend/CRM for attribution tracking.
### Phase 7: Validation and Delivery
#### 7.1 Test in Browser
Open dev server URL and verify:
- All pages load without errors
- Navigation works (header menu, breadcrumbs)
- Forms submit correctly
- Mobile responsive (test sticky CTA buttons)
- Cookie banner appears on first visit
- Images load correctly
#### 7.2 SEO Validation
Verify against `references/seo-checklist.md`:
- Unique title and description on each page
- H1 hierarchy correct
- Images have alt attributes
- robots.txt and sitemap.xml accessible
- Structured data present on key pages
#### 7.3 Create Checkpoint
```bash
webdev_save_checkpoint "Complete lead-gen website - [X] pages, SEO optimized, RGPD compliant"
```
#### 7.4 Deliver to User
Send checkpoint attachment via `message` tool with:
- Summary of what was built
- Page count and key features
- SEO optimizations implemented
- Next steps (publish, custom domain, Google Search Console)
## Bundled Resources
### Scripts
**`scripts/generate_pages_batch.py`**
Generate multiple similar pages from template and data file.
Usage: `python generate_pages_batch.py <template> <data_json> <output_dir>`
**`scripts/create_seo_files.py`**
Generate robots.txt and sitemap.xml automatically.
Usage: `python create_seo_files.py <domain> <pages_json> <output_dir>`
**`scripts/generate_content_structure.py`**
Create content structure markdown from specifications JSON.
Usage: `python generate_content_structure.py <specs_json> <output_md>`
### Templates
**Components:**
- `component-Header.tsx` - Sticky header with logo, nav, CTA
- `component-Footer.tsx` - Footer with links and contact info
- `component-SEOHead.tsx` - SEO meta tags and structured data
- `component-Breadcrumbs.tsx` - Navigation breadcrumbs
- `component-ContactForm.tsx` - Form with UTM tracking
- `component-CookieBanner.tsx` - RGPD cookie consent banner
**Pages:**
- `page-service-template.tsx` - Service page template
- `page-legal-template.tsx` - Legal page template
- `design-ideas-template.md` - Design brainstorming structure
### References
**`references/seo-checklist.md`**
Complete SEO checklist covering meta tags, structured data, technical SEO, on-page SEO, local SEO, and content quality. Read this before Phase 6 to ensure nothing is missed.
**`references/conversion-best-practices.md`**
Best practices for maximizing conversions: CTA strategy, contact options, trust signals, form optimization, mobile optimization. Consult during Phase 5 when building pages.
**`references/rgpd-compliance.md`**
Complete RGPD compliance guide covering cookie banner, privacy policy, cookie policy, legal mentions, forms, consent, data security, and user rights. Essential for Phase 6.
**`references/design-philosophies.md`**
Five example design philosophies (Neo-Artisanat Digital, Brutalist Confidence, Soft Modernism, Vibrant Energy, Luxury Minimalism) with selection criteria. Use as inspiration during Phase 2.
## Tips and Best Practices
**Design consistency:** Document chosen design philosophy at the top of each CSS/component file as a reminder.
**Image optimization:** All images should be stored in `/home/ubuntu/webdev-static-assets/` and referenced via CDN URLs to avoid deployment timeouts.
**Content quality over quantity:** Better to have 10 excellent pages than 20 mediocre ones. Focus on answering user intent.
**Mobile-first:** Design and test mobile experience first. Most local service searches happen on mobile.
**Conversion priority:** Every page should have clear CTAs. Phone and WhatsApp buttons should be always visible on mobile.
**Local SEO:** Mention city/region name in titles, H1, and content. Create separate pages for each service area if covering multiple cities.
**Fast iteration:** Use batch generation scripts for similar pages to save time. Focus manual effort on unique, high-value pages.
**Testing:** Always test in browser before creating checkpoint. Check mobile responsive, form submission, and navigation.
## Common Pitfalls
**Skipping design brainstorming:** Leads to generic, forgettable designs. Always create ideas.md with 3 distinct approaches.
**Generating images during development:** Inefficient. Generate all images upfront in Phase 3.
**Weak content:** Thin content (<300 words) won't rank. Invest time in Phase 4 to create substantial, helpful content.
**Missing RGPD elements:** Cookie banner, privacy policy, and legal mentions are REQUIRED in EU. Don't skip Phase 6.3.
**No UTM tracking:** Without UTM parameters, you can't measure campaign effectiveness. Ensure ContactForm captures them.
**Forgetting mobile CTAs:** Desktop-only CTAs lose mobile conversions. Always add sticky mobile buttons.
**Creating checkpoint during development:** Only create ONE checkpoint at the end (Phase 7). Multiple checkpoints confuse users during initial delivery.
用 gh 命令行工具查看 GitHub 上的 PR 检查、工作流运行与 API 查询的技能。
聚合 16 个搜索引擎、无需 API Key 的检索技能。
用命令行直接列出、配置、授权并调用 MCP 服务与工具。
生成有效的 Mermaid 图表,如流程图、时序图、思维导图、ER 图、用户旅程等。
用受管 OAuth 访问 YouTube Data API,检索视频并管理频道内容。
通过受管的 OAuth 连接访问 GitHub REST API。
覆盖规划、实现、验证与测试的编码工作流技能。
整理 Docker 常用命令与容器工作流的技能。
数据月份:2026-09 · 下载/星标/安装数聚合自公开技能注册库(ClawHub、SkillHub),按月更新。口径说明见数据方法。