Paste any URL and get an instant Agent Readiness Score — Auspia checks 18 AI agent standards including MCP, A2A, OAuth discovery, robots.txt, and emerging agent commerce protocols.
Powered by 18+ protocol checks · No signup required · Results in seconds
LEADERBOARD →Agent Readiness measures how well your website can be discovered, read, and acted upon by AI agents — ChatGPT, Claude, Perplexity, Gemini, and the next generation of autonomous software. It's the foundation of AEO (Answer Engine Optimization): the practice of making your site legible and actionable for AI, not just humans.
Can AI agents find your site's structure, sitemap, and entry points through standard protocols?
Can agents read your content in machine-friendly formats like Markdown, and understand your bot access rules?
Can agents authenticate, call your APIs, use your MCP server, and complete transactions on your site?
Auspia audits your site across five categories that cover the full agent interaction stack — from basic crawlability to real-time agent transactions.
MCP server cards, A2A agent cards, agent skills indexes, API catalogs, and OAuth discovery endpoints.
Payment and commerce protocol support for agent-initiated transactions: x402, MPP, UCP, ACP, AP2.
robots.txt presence, XML sitemap, and RFC 8288 Link headers for agent-friendly navigation.
Markdown content negotiation so agents can request clean, structured text instead of raw HTML.
AI-specific crawler rules, Web Bot Auth, and content signals that tell agents what they can use.
ChatGPT, Claude, Perplexity, and Gemini are already browsing the web on behalf of millions of users. Sites that aren't agent-ready get skipped — or worse, misrepresented. Your Agent Readiness Score tells you exactly where you stand.
Answer Engine Optimization (AEO) is how forward-thinking teams are preparing for the AI-first web. Just as SEO made sites findable by search engines, AEO makes sites usable by AI agents. Your Agent Readiness Ranking is your AEO baseline.
Every failed check comes with an AI-generated fix prompt you can paste directly into Claude Code or Cursor. Improve your score, improve your agent readiness ranking, and stay ahead of the AEO curve.
Enter any public URL — your homepage, a product page, or your API docs. No account required.
Auspia runs 18 checks across 5 categories and returns a score from 0–100 with a readiness level from 0 (Not Ready) to 5 (Agent-Native).
Every failed check includes a ready-to-use fix prompt. Paste it into Claude Code, Cursor, or any AI coding tool to implement the fix in minutes.
Free, instant, no signup. See exactly where your site stands on the AEO readiness scale.
An Agent Readiness Score is a 0–100 rating that measures how well your website supports AI agents. Auspia checks 18 standards across five categories — discoverability, content accessibility, bot access control, agent discovery, and agent commerce — and converts the results into a single score with a readiness level from 0 (Not Ready) to 5 (Agent-Native).
Each of the 18 checks returns pass, fail, or neutral. Passed checks contribute to your score; failed checks reduce it. Neutral checks (like agent commerce protocols on a non-commerce site) don't count against you. The final score is a weighted percentage across all applicable checks.
AEO stands for Answer Engine Optimization — the practice of making your site legible and actionable for AI agents, not just search engines. Your Agent Readiness Score is your AEO baseline. A higher score means AI agents like ChatGPT, Claude, and Perplexity can better discover, read, and act on your content.
Auspia checks 18 standards: robots.txt, sitemap, Link headers, Markdown content negotiation, AI crawler rules, Web Bot Auth, content signals, MCP server card, A2A agent card, agent skills, API catalog, OAuth discovery, OAuth protected resource, Web MCP, x402, MPP, UCP, ACP, and AP2.
Yes — the Agent Readiness Scanner is completely free. No account required, no rate limits for normal use. Just paste a URL and get your score in seconds.
Every failed check in your report includes an AI-generated fix prompt. Paste it into Claude Code, Cursor, or any AI coding assistant to implement the fix. Start with the checks required for your next readiness level — the 'How to reach the next level' section shows exactly what to fix first.
Level 0 Not Ready · 1 Basic Web Presence · 2 Bot-Aware · 3 Agent-Readable · 4 Agent-Actionable · 5 Agent-Native. Each level adds a tier of capabilities, from publishing robots.txt all the way to running an MCP server that agents can act on.
Scans are submitted as Unlisted, so they aren't published in public scan listings. The full evidence is visible to anyone who has the scan ID, but your URL won't appear in any public index.