Lead Gen Website Builder (SEO PUR)

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.

Security Scan

The skill's files and runtime instructions are coherent with its stated purpose (building local lead‑generation websites); nothing in the bundle requests unrelated credentials or network exfiltration, but there are a few operational mismatches you should be aware of before running it.

Purpose & Capability

The code files (React templates, cookie banner, SEO head, contact form) plus Python scripts (sitemap/robots, content structure, batch page generation) and the numerous reference documents align with the declared purpose: building SEO‑optimized local lead‑gen websites with RGPD guidance and tracking. Templates and automation are proportional to the stated workflow.

Instruction Scope

SKILL.md instructs the agent to run Python scripts and to use external helpers: a `generate` tool for images and `webdev_init_project`. It also specifies absolute host paths (e.g., /home/ubuntu/webdev-static-assets/ and /home/ubuntu/skills/lead-gen-website/). The package metadata did not declare required binaries or tools (python, image generator, webdev_init_project), so the runtime relies on the agent environment having these tools and filesystem write access. This is expected for a development skill but should be verified: the skill will read/write local files and expects the ability to create assets and pages on the host filesystem.

Install Mechanism

There is no install spec — this is instruction‑plus‑templates with simple Python scripts. That is lower risk than arbitrary remote installs. The included scripts are plain (no remote downloads, no obfuscated code) and write local files only.

Credentials

The skill requests no environment variables or credentials. It includes placeholders for analytics (GTM/GA4) and advises adding tracking IDs, but it does not require or attempt to read secrets. No unexpected credentials or unrelated service tokens are requested.

Persistence & Privilege

always:false and no install hooks are present. The skill reads/writes its own templates and output files but does not request permanent platform privileges or modify other skills' configurations.

Assessment

What to check before installing/running - Verify environment/tools: The instructions expect Python and two external tools referenced as `generate` (image generator) and `webdev_init_project`. Confirm those binaries/tools exist and are trusted in the agent environment before running. - File writes: The workflow writes files under absolute paths (/home/ubuntu/...), creates pages, images, robots/sitemap, etc. Run in a controlled workspace (not a production host) if you are unsure what will be written. - Image generation: The skill directs generation of images and storing them locally. Confirm the `generate` tool's provider and privacy policy if you care about sensitive data or copyright in generated images. - Analytics and third parties: The templates mention GTM/GA4, WhatsApp links, Google Fonts and external endpoints (fonts.googleapis.com, wa.me). These are expected for this use case but will cause network calls when the site is live; supply your own analytics IDs and ensure you only enable analytics after cookie consent as suggested. - GDPR/transparency: The skill includes guidance to disclose 'mise en relation' and avoid doorway/scaled spam pages. If you use the skill to build sites for others, ensure you follow the guardrails (real addresses/photos, honest claims) to avoid legal or search‑engine penalties. - Safety posture: The code appears readable and non‑obfuscated; scripts do not contact remote servers or exfiltrate data. If you need higher assurance, run the Python scripts with a dry‑run or in an isolated container and review generated files before publishing. If you want, I can list the exact commands the SKILL.md expects or extract all absolute paths and external endpoints referenced so you can review them before executing.

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:

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

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:

<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:

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:

<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:

[
  {"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:

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):

const jsonLd = {
  "@context": "https://schema.org",
  "@type": "LocalBusiness",
  "name": "Business Name",
  "telephone": "[phone]",
  "email": "[e-mail]",
  "address": {
    "@type": "PostalAddress",
    "addressLocality": "City",
    "addressCountry": "FR"
  },
  "areaServed": ["City1", "City2"],
  "openingHours": "Mo-Fr 08:00-18:00"
};

Service pages (Service):

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

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.