Is your site ready for agents?
Paste a URL. It probes the site the way an agent would — looking for the surfaces an agent uses to discover and call your product (llms.txt, AGENTS.md, an MCP server card, OpenAPI, OAuth discovery, markdown content negotiation) — and scores it 0–100 against the Agents First framework. You get a shareable result card in a few seconds. No login.
Probes the site for agent-discoverable surfaces — llms.txt, AGENTS.md, MCP card, OpenAPI, OAuth, markdown. No login.
What it checks
- Discoverability — does
robots.txtaddress AI agents, is there an/llms.txt, an/AGENTS.md? - Content accessibility — markdown content negotiation, sitemap, a discoverable API catalog.
- Bot access control — a deliberate AI policy vs. blanket allow/deny.
- Agent capabilities — a published MCP server card, a CLI/SDK, OAuth-with-PKCE discovery.
- Visibility — do your docs tell agents how to install your MCP server or use your CLI?
It scores what’s live at the URL you give it — so the apex of a big site can score lower than a hand-reviewed deep-dive. That’s the point: it’s what an agent finds when it lands.
See example scores or read the eight principles behind the rubric.