Free diagnostic
Is your website ready for AI agents?
ChatGPT, Claude, and Perplexity are already trying to crawl, read, and cite your site. Scan any URL for the checks that decide whether they can — discoverability, bot access rules, and MCP/agent discovery — and get exact fixes for what's failing.
Enter your website
Why it matters
Search is turning into a conversation your website isn't part of
Agents can't click through a JavaScript-only nav or guess at an undocumented API. Sites that publish clear, machine-readable rules for crawling, content, and tool access get read, cited, and transacted with — sites that don't get skipped.
This isn't a subjective grade — every check maps to a published, open specification: the same rules AI crawlers and agents actually read, and the ones companies like Cloudflare, OpenAI, and Anthropic are converging on as the infrastructure for an agentic web.
Discoverability
Can search engines and AI crawlers find your site at all?
Based on robots.txt, sitemaps, and DNS-AID service records
Content Accessibility
Can an agent actually read your content, not just a browser's rendered version of it?
Based on Markdown content negotiation for pages that are JavaScript-heavy or noisy in raw HTML
Bot Access Control
Have you told AI crawlers what they’re allowed to do?
Based on AI-specific robots.txt directives (GPTBot, ClaudeBot, PerplexityBot, and others) and the Content Signals standard
API, Auth & MCP Discovery
Can an agent find and use your tools and APIs on its own?
Based on Model Context Protocol (MCP) server cards, Agent2Agent (A2A) agent cards, and OAuth/auth.md discovery
Commerce Protocols
Can an agent complete a purchase on your behalf?
Based on emerging agentic-commerce standards like x402 and the Agentic Commerce Protocol
Failing checks?