AEO MCP Server

The AEO.dev Model Context Protocol (MCP) server integrates AEO analysis directly into AI assistants like Claude. Get AEO recommendations while you code.

What is MCP?

The Model Context Protocol is a standard developed by Anthropic for connecting AI assistants to external tools and data sources. With our MCP server, Claude and other MCP-compatible AI can:

• Analyze your content for AEO readiness
• Suggest structured data improvements
• Generate Schema.org markup
• Validate llms.txt files

Installation

npm install -g @aeo/mcp-server

Configuration

Add to your Claude desktop configuration:

{
  "mcpServers": {
    "aeo": {
      "command": "aeo-mcp-server",
      "args": []
    }
  }
}

Available Tools

analyze_url

Analyze a URL for AEO readiness:

Tool: analyze_url
Input: { "url": "https://yoursite.com/page" }

Output:
{
  "score": 75,
  "breakdown": {
    "content_structure": 80,
    "structured_data": 65,
    "machine_readability": 85,
    "authority_signals": 70
  },
  "recommendations": [...]
}

generate_schema

Generate Schema.org markup for content:

Tool: generate_schema
Input: {
  "type": "Article",
  "content": {
    "headline": "Article Title",
    "description": "Article description",
    "author": "Author Name"
  }
}

Output:
{
  "@context": "https://schema.org",
  "@type": "Article",
  "headline": "Article Title",
  ...
}

validate_llms_txt

Validate an llms.txt file:

Tool: validate_llms_txt
Input: { "url": "https://yoursite.com/llms.txt" }

Output:
{
  "valid": true,
  "warnings": [
    "Consider adding 'topics' field for better AI understanding"
  ],
  "suggestions": [...]
}

check_robots

Check robots.txt for AI crawler access:

Tool: check_robots
Input: { "url": "https://yoursite.com" }

Output:
{
  "ai_crawlers": {
    "GPTBot": "allowed",
    "Google-Extended": "allowed",
    "PerplexityBot": "blocked",
    "ClaudeBot": "allowed"
  },
  "recommendations": [...]
}

Usage with Claude

Once configured, ask Claude to analyze your AEO:

"Can you analyze the AEO readiness of my homepage at https://mysite.com?"

Claude will use the MCP tools to:

1. Fetch and analyze your page
2. Check structured data implementation
3. Validate llms.txt and robots.txt
4. Provide recommendations

Example Conversation

You: Can you check if my article page is optimized for AI search?

Claude: I'll analyze your page for AEO readiness.

[Uses analyze_url tool]

Your page scores 68/100 for AEO. Here's the breakdown:

• Content Structure: 75/100 ✓ Good heading hierarchy
• Structured Data: 45/100 ⚠️ Missing Article schema
• Machine Readability: 80/100 ✓ llms.txt present
• Authority Signals: 72/100 ⚠️ Author info incomplete

Recommendations:

1. Add Article schema with JSON-LD
2. Include author credentials and photo
3. Add dateModified to existing metadata

Would you like me to generate the Schema.org markup?

Development Integration

Use the MCP server in your CI/CD pipeline:

# .github/workflows/aeo-check.yml
name: AEO Analysis
on: [push]

jobs:
  analyze:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3
      - name: Install AEO MCP Server
        run: npm install -g @aeo/mcp-server
      - name: Analyze
        run: aeo-analyze --url ${{ env.PREVIEW_URL }}

Resources

Available Resources

The MCP server exposes these resources:

Next Steps