Agent-Ready Checker

Can AI assistants and agents find, read, and use a business website? More customers now ask ChatGPT, Claude, Perplexity or Google's AI where to go before they ever search. This Actor checks whether those assistants can actually read a site, and tells you what to fix.

Give it one or more URLs. For each site you get a score out of 100 and prioritized findings, each with a plain-language fix.

What it checks

• AI access in robots.txt: whether ChatGPT (GPTBot, OAI-SearchBot, ChatGPT-User), Claude (ClaudeBot, Claude-User), Perplexity, Google AI (Google-Extended) and Apple AI are allowed or blocked
• Structured business data: schema.org JSON-LD for the business (name, address, phone, hours, website)
• Summary text: page title and meta description that assistants can quote
• Readable content: real text on the page, not a JavaScript-only shell
• Contact and action paths: clickable phone or email, and a booking, order or quote path
• Discovery files: sitemap.xml and llms.txt

It reads only public pages a crawler would read (home page, robots.txt, llms.txt, sitemap.xml). It is deterministic and fast, and makes no AI calls. No logins, no personal data.

Who it's for

• AI agents and automations that need a quick, structured readiness check on a business website before recommending, booking or integrating with it
• Agencies, SEO and web professionals auditing client sites for AI search ("AEO/GEO")
• Lead-gen and sales workflows: score a list of local businesses and find the ones that need help
• Business owners who want to know whether AI assistants can find them

Input

{ "urls": ["https://example.com", "https://another-business.com"] }

Up to 50 URLs per run.

Output (one record per URL)

Real example (checking omenz.org):

{
  "url": "https://omenz.org",
  "score": 59,
  "findings": [
    {
      "severity": "high",
      "code": "no_structured_data",
      "what_it_means": "No structured data (schema.org JSON-LD) found, so assistants have to guess your name, address, hours and services.",
      "fix": "Add a LocalBusiness (or Organization) JSON-LD block with name, address, phone, hours and website."
    },
    {
      "severity": "medium",
      "code": "no_meta_description",
      "what_it_means": "No meta description, so assistants have no short summary to quote.",
      "fix": "Add a one-sentence description: what you do, for whom, and where."
    }
  ],
  "facts": {
    "https": true, "robots_found": false, "ai_bots_blocked": [],
    "business_schema": false, "llms_txt": false, "sitemap": false,
    "phone_link": false, "email_link": false, "booking_path": false
  }
}

Findings are sorted by severity (high, medium, low). Every finding has a stable code, so agents and scripts can act on it.

Pricing

Pay per result: $0.10 per website checked. No subscription.

Using it from an agent or script

Call it like any Apify Actor: through the Apify API, the Apify client libraries, or Apify's integrations for AI agents. Pass urls and read the dataset items.

Limits and honesty

• It checks readiness signals; it does not guarantee rankings, traffic or how any AI assistant will answer.
• llms.txt is an emerging convention, so it's scored low priority, not as a requirement.
• Sites that block automated requests may return an error for that URL.

Tell us what you need

Using this in an agent, an automation or a client workflow? Open an issue on this Actor's Issues tab (https://apify.com/omenz/agent-ready-checker/issues) and tell us:

• what you're using it for
• what you'd want checked next
• whether you'd pay for a version that also generates the fixes (structured data, summary, llms.txt)

We read every one. If it's working for you, a quick review helps other people find it.

Want the fixes done for you?

OMENZ also runs a human-reviewed Agent-Ready Business Audit: live AI-assistant answer checks plus a done-for-you fix package.

Built by OMENZ: https://omenz.org