Audit your site for AI readiness from your assistant

Connect it in one step

Pick your client. The server needs no key; every snippet below works as it is.

claude mcp add --transport http nodpick https://nodpick.com/mcp

https://nodpick.com/mcp

{
  "mcpServers": {
    "nodpick": {
      "url": "https://nodpick.com/mcp"
    }
  }
}

{
  "servers": {
    "nodpick": {
      "type": "http",
      "url": "https://nodpick.com/mcp"
    }
  }
}

https://nodpick.com/mcp

URL:        https://nodpick.com/mcp
Transport:  Streamable HTTP (stateless, no session)
Auth:       none (optional header: Authorization: Bearer npk_...)

claude mcp add --transport http nodpick https://nodpick.com/mcp --header "Authorization: Bearer npk_your_key"

{
  "inputs": [
    { "type": "promptString", "id": "nodpick-key", "description": "Nodpick API key", "password": true }
  ],
  "servers": {
    "nodpick": {
      "type": "http",
      "url": "https://nodpick.com/mcp",
      "headers": { "Authorization": "Bearer ${input:nodpick-key}" }
    }
  }
}

The tools

Five tools work for everyone. Two more appear when you connect with an API key. All of them are read-only.

check is one of crawler_access, training_crawlers, raw_html_content, indexability, metadata, structured_data, freshness, agent_extras. framework is one of nextjs, astro, wordpress, static, spa, nginx, apache, other; leave it out and the audit detects the stack.

An example run

Ask your assistant, for example: "Audit example.com for AI readiness and fix the most important issue." It typically calls the tools in this order.

1. 1

   audit_ai_readiness
   { "url": "https://example.com", "framework": "nextjs" }
   -> score 0-100, weighted checks, top fixes with fix.snippetId

2. 2

   get_fix_snippet
   { "check": "crawler_access", "framework": "nextjs" }
   -> copy-paste code for that check on that stack

3. 3

   get_free_visibility_check_link
   { "domain": "example.com", "country": "DE", "language": "en" }
   -> https://nodpick.com/check?domain=example.com&ref=mcp&country=DE&language=en

The audit shows whether AI crawlers can read your site. Whether assistants actually name your brand is a different question, and the free visibility check answers it.

Prompts to try

Copy one into your assistant once the server is connected. Replace example.com with your site.

• Audit and first fix

  Copy

  Audit https://example.com for AI readiness. The site runs on Next.js. List the three fixes that gain the most points and give me the code for the first one.

  audit_ai_readinessget_fix_snippet

• Can AI crawlers get in?

  Copy

  Check whether GPTBot, ClaudeBot, PerplexityBot and Google's crawlers may fetch https://example.com/pricing, and tell me which robots.txt rule blocks any of them.

  check_ai_crawlers

• llms.txt

  Copy

  Does https://example.com have a valid llms.txt? If not, draft one for my site from the starter template.

  check_llms_txt

• Fix it in this repo

  Copy

  Audit https://example.com, then apply the fix for the check that gains the most points directly in this codebase (framework: astro).

  audit_ai_readinessget_fix_snippet

• After a robots.txt change

  Copy

  I just changed robots.txt on https://example.com. Check it again and tell me whether any AI search crawler lost access.

  check_ai_crawlers

• Do assistants name my brand?

  Copy

  Give me the free visibility check link for example.com in Germany, in English.

  get_free_visibility_check_link

• My action planAPI key

  Copy

  List my tracked brands and show the action plan for the first one, starting with the highest-priority step.

  list_projectsget_action_plan