Landing Gen

Generate landing page content from PRD — hero section, features, A/B headline variants, CTA, and SEO meta tags. Use when user says "create landing page", "wr...

Security Scan

The skill's requested capabilities and instructions align with its stated purpose of generating landing content and optionally scaffolding pages; it reads project docs and writes output files but does not ask for unrelated credentials or installs.

Purpose & Capability

Name and description (generate landing content and scaffold pages) match the instructions: read PRD/README, detect stack, produce content, and write markdown or astro/Next files. It does not request unrelated env vars or external credentials.

Instruction Scope

Instructions limit actions to project files (README, docs/prd.md, research.md), optional web competitor analysis (via web_search), and writing scaffolded pages or docs/landing-content.md. However, allowed-tools include Bash, Write, and Edit, which permit arbitrary file modification and shell commands; the instructions themselves don't instruct exfiltration or reading unrelated system files, but those broad tool permissions mean a user should expect the skill can modify repo files and run shell commands if invoked autonomously.

Install Mechanism

No install spec and no code files — instruction-only. Nothing is downloaded or written to disk by an installer beyond normal Write/Edit actions performed at runtime.

Credentials

The skill requests no environment variables, credentials, or config paths. The data it reads (PRD/README/research.md) is consistent with its stated goal.

Persistence & Privilege

always:false (normal). The skill is allowed to run autonomously and has Write/Edit/Bash tool permissions, so it can create or modify project files when invoked — this is expected for a scaffolding/content generator, but grants the ability to change files and run shell commands, so review outputs and history after use.

Assessment

This skill appears coherent for generating landing copy and scaffolding pages. Before installing or running it: (1) provide a clear PRD/README so outputs are high quality; (2) be aware the skill can write files (docs/landing-content.md or src/pages/index.astro) and may run shell commands because it lists Bash/Write/Edit as allowed tools — review any created/modified files and commits before publishing; (3) avoid storing secrets in the project docs it will read (PRD/README) since the skill reads these files; (4) if you want to limit risk, run it in a cloned or sandboxed repo, or deny shell execution while allowing read/write of specific paths; (5) if you need network competitor analysis disabled, avoid granting the web_search tool or provide competitor info yourself.

Generate landing page content from a project's PRD. Produces hero section, features, social proof, CTA, SEO meta tags, and A/B headline variants. If astro-static stack detected, can scaffold actual page files.

MCP Tools (use if available)

• kb_search(query) — find related methodology (conversion, copywriting)
• project_info(name) — get project stack and details
• web_search(query) — competitor landing analysis

If MCP tools are not available, fall back to Glob + Grep + Read.

Steps

1. Parse project from $ARGUMENTS.

   • Read PRD, README, or research.md for product info.
   • If empty: ask via AskUserQuestion.

2. Detect stack:

   • Check for astro.config.* → astro-static (can scaffold page)
   • Check for next.config.* → Next.js (can scaffold route)
   • Otherwise: generate content-only markdown

3. Extract landing inputs from PRD/README:

   • Problem: 1 sentence pain statement
   • Solution: 1 sentence product description
   • ICP: Target user persona
   • Features: Top 3-4 differentiating features with descriptions
   • Competitors: From research.md (if exists) — for positioning
   • Pricing: If available

4. Competitor landing analysis (optional, if MCP/WebSearch available):

   • Search for top 3 competitor landing pages
   • Note: headline patterns, CTA language, social proof types
   • Identify positioning gaps

5. Forced reasoning — conversion strategy: Before generating, write out:

   • Primary conversion: What's the ONE action? (sign up / download / buy)
   • Objections: Top 3 reasons someone wouldn't convert
   • Trust signals: What overcomes each objection?
   • Above the fold: Problem + Solution + CTA — nothing else

6. Generate landing content:

   6a. Hero Section

   • Headline: Problem-focused, benefit-driven (8-12 words)
   • Subheadline: How the product solves it (15-25 words)
   • CTA button: Action verb + outcome ("Start Free Trial", "Download Now")
   • Visual: Describe what image/screenshot/demo should be shown

   6b. A/B Headline Variants (5 options)

   Generate 5 distinct headline approaches:

   1. Pain-focused: "Tired of {problem}?"
   2. Benefit-focused: "{Outcome} without {hassle}"
   3. Curiosity: "The {adjective} way to {action}"
   4. Social proof: "Join {N}+ {users} who {outcome}"
   5. Direct: "{Product}: {one-line value prop}"

   6c. Features Section (3-4 features)

   For each feature:

   • Icon suggestion (emoji or icon name)
   • Feature title (3-5 words)
   • Feature description (1-2 sentences)
   • Benefit framing (what user gains, not what product does)

   6d. Social Proof Section

   • Placeholder for testimonials (template with name, role, quote)
   • Metrics placeholders ("X+ users", "Y% faster", "Z countries")
   • Trust badges placeholder (App Store rating, awards, press)

   6e. CTA Section

   • Repeat headline (or variation)
   • CTA button (same as hero)
   • Risk reducer ("Free forever" / "No credit card" / "Cancel anytime")

   6f. SEO Meta Tags

   <title>{title — 50-60 chars}</title>
   <meta name="description" content="{description — 150-160 chars}">
   <meta property="og:title" content="{og title}">
   <meta property="og:description" content="{og description}">
   <meta property="og:image" content="{suggest image dimensions 1200x630}">
   <meta property="og:type" content="website">
   

7. Write output:

   If astro-static detected: scaffold page at src/pages/index.astro or new route.

   Otherwise: write to docs/landing-content.md:

   # Landing Page Content: {Project Name}
   
   **Generated:** {YYYY-MM-DD}
   **Primary CTA:** {action}
   **Target ICP:** {persona}
   
   ## Hero
   **Headline:** {headline}
   **Subheadline:** {subheadline}
   **CTA:** {button text}
   **Visual:** {description}
   
   ## A/B Headlines
   1. {variant 1}
   2. {variant 2}
   3. {variant 3}
   4. {variant 4}
   5. {variant 5}
   
   ## Features
   {features with icons and descriptions}
   
   ## Social Proof
   {templates and placeholders}
   
   ## Final CTA
   {closing section}
   
   ## SEO Meta Tags
   {html meta tags}
   
   ---
   *Generated by /landing-gen. Fill in social proof, add visuals, then publish.*
   

8. Output summary — headline, CTA, and suggested next steps.

Notes

• Headline should be testable — run A/B with 2-3 variants
• "Above the fold" = hero only — don't overload
• Social proof is placeholder — fill with real data as it comes
• For fake-door tests: hero + CTA + email capture is enough (2-hour launch)
• Works with /seo-audit — generate content, then audit the deployed page

Common Issues

No PRD or product info found

Cause: Project lacks docs/prd.md or README with product description. Fix: Run /validate to generate PRD first, or provide a README with problem/solution/features.

Headlines too generic

Cause: Weak problem statement or missing competitor differentiation. Fix: Add research.md with competitive analysis. Specific pain points produce specific headlines.

Stack not detected for page scaffolding

Cause: No astro.config.* or next.config.* found. Fix: Skill outputs content-only markdown by default. To get actual page files, ensure the project uses astro-static or nextjs-supabase stack.