NuxtSEO Pro
Mcp

AI Content Generation Prompts

Pre-built prompts that chain multiple MCP tools into complete workflows. Create content briefs, generate articles, or run full page SEO improvements—your AI handles the orchestration.

improve_page_seo

Analyze a Nuxt page and generate SEO improvements. Orchestrates multiple tools in sequence.

ParameterTypeDescription
filePathstringPath to Vue file (e.g., app/pages/about.vue)
fileContentstringContent of the Vue file
topicstringPage topic for keyword research (optional, auto-detected)
nuxtConfigstringnuxt.config.ts content (optional, for module detection)
liveUrlstringLive URL to extract rendered meta (optional, compares source vs output)
liveHtmlstringRaw HTML from local dev server (optional, for non-deployed pages)
// Basic usage
improve_page_seo({
  filePath: 'app/pages/tools/meta-checker.vue',
  fileContent: '...',
  topic: 'meta tag checker tool'
})

// With live comparison
improve_page_seo({
  filePath: 'app/pages/tools/meta-checker.vue',
  fileContent: '...',
  liveUrl: 'https://mysite.com/tools/meta-checker'
})

// With local dev HTML
improve_page_seo({
  filePath: 'app/pages/tools/meta-checker.vue',
  fileContent: '...',
  liveHtml: '<html>...</html>' // from $fetch('http://localhost:3000/tools/meta-checker')
})

Workflow

The prompt runs these tools in sequence:

  1. check_meta_tags (optional) - Extract rendered meta from live URL (via free MCP)
  2. analyze_page - Detect page type, existing SEO, generate suggestions
  3. research_keywords - Find long-tail keywords (may return empty for niche technical topics)
  4. generate_schema_org - Generate structured data if missing
  5. generate_og_image_template - Create OG image component if missing

Output

Returns:

  • Rendered vs Source (if liveUrl/liveHtml provided) - Comparison of what's in source vs what actually renders
  • Top 5 long-tail keywords with volume/difficulty
  • Current SEO status (what already exists)
  • Ready-to-paste code blocks:
    • useSeoMeta() with researched keywords
    • useSchemaOrg() for the page type
    • defineOgImage() configuration
  • Implementation notes (modules to install)

Content Workflow

The content prompts work together:

research_keywords → content_brief → article_generation
  1. Research keywords - Find target keywords for topic
  2. Create brief - Generate structured outline
  3. Write article - Generate full content following the style guide

content_brief

Generate a structured outline for an article. Use after keyword research.

ParameterTypeDescription
topicstringMain topic for the article
targetKeywordsstringComma-separated target keywords (3-5)
competitorUrlsstringComma-separated URLs to differentiate from (optional)
pageTypestringtechnical, marketing, or tutorial (default: technical)
content_brief({
  topic: 'Adding Schema.org to Nuxt Pages',
  targetKeywords: 'nuxt schema.org, useSchemaOrg nuxt, nuxt structured data',
  pageType: 'technical'
})

Output Structure

The prompt returns:

  1. Primary search intent - What the user wants to accomplish
  2. Unique angle - How to differentiate from competitors
  3. H2 sections (4-6) - Each with:
    • Heading text
    • Bullet points to cover
    • Target keywords to use
  4. Key questions - Questions users ask about the topic
  5. Frontmatter suggestion - Title and related pages

Rules Applied

The brief follows the Writing Guide:

  • Code within first 3 scrolls
  • Parameters in tables
  • Primary keyword in H1 and first paragraph
  • H2s as useful navigation, not keyword lists
  • Some H2s as questions users ask

article_generation

Generate a full article from an outline. Embeds the complete Writing Guide rules.

ParameterTypeDefaultDescription
outlinestringrequiredContent brief from content_brief
targetWordCountstring"1500"Target length (500-5000)
includeCodeExamplesstring"true"Include code examples (true/false)
sitemapUrlsstring-Comma-separated site URLs for internal linking
article_generation({
  outline: '# Content Brief\n\n## Primary Intent...',
  targetWordCount: '2000',
  includeCodeExamples: 'true',
  sitemapUrls: '/docs/schema-org/getting-started, /docs/seo-utils/api/use-seo-meta'
})

Embedded Rules

The prompt includes:

Banned words: dive into, crucial, leverage, ensure, comprehensive...

Banned phrases: "it's important to note", "in today's X", "let's explore"...

Quick fixes:

SlopFix
It's important to note...just state it
This allows you to...You can...
In order to...To...

Voice rules:

  • Developer-to-developer, casual but accurate
  • State opinions: "This is lazy", "overkill for most sites"
  • Say what NOT to do

Structure rules:

  • Code within first 3 scrolls
  • Parameters in tables, not prose
  • No "Best practices" sections

Gap Markers

The output includes markers for content that needs follow-up:

[STAT NEEDED: percentage of sites with broken meta tags]
[VERIFY: does this work in Nuxt 4?]
[EXAMPLE NEEDED: real-world product schema]
[LINK: internal link to related page]

Review the article and fill these gaps before publishing.

Chaining Prompts

A full workflow for creating a new article:

// 1. Research keywords
const keywords = await research_keywords({ topic: 'nuxt meta tags' })

// 2. Create outline
const brief = await content_brief({
  topic: 'Adding Meta Tags in Nuxt',
  targetKeywords: keywords.keywords.map(k => k.keyword).slice(0, 5).join(', ')
})

// 3. Generate article
const article = await article_generation({
  outline: brief,
  targetWordCount: '1500'
})
Edit this page
Markdown For LLMs
Did this page help you?

Keyword Research

MCP tools for keyword research and SERP analysis. Find long-tail keywords, analyze competition, and discover content gaps—through your AI assistant.

Social Signals

MCP tool for analyzing social proof and community activity. Validate market demand across GitHub, Reddit, and Twitter before building.