Agent Readiness Compare

How to choose an agent readiness scanner: eight questions to ask

A buyer's guide for web and growth leads comparing tools that score a site for AI agents.

In short

Ask eight questions: which standards the scanner checks, whether it runs an agent or only reads files, whether its scoring rules are published, whether you can run it from code, what fix path it offers, what access it needs, what it cannot see, and how fast it follows changes. Then run two scanners on the same pages.

Scanners that look alike give different grades

Every scanner takes a URL and returns a score. Underneath, they differ in what they check, how they weight it and what they can reach. Two scanners can grade the same site differently for defensible reasons. Know those differences before you pick one, and before you report a number to leadership.

The eight questions below follow the criteria on the method page. They apply to any tool in the category. The table shows what the six scanners profiled here publish.

What six scanners publish about checks, scoring and access, read on 2026-11-04
ScannerChecksScoring rulesAccess
Ora132 in the API catalog, 126 on the methodology pageFour layers weighted 20, 30, 40 and 10; grade bands A+ to FFree; anonymous scans capped at 30 per IP per day; REST API, MCP server, CLI
Cloudflare isitagentready.com23 in five categoriesReadiness level 0 to 5; no check weights publishedFree, no sign-up; scan API and MCP endpoint; URL Scanner needs an account and token
is-agentic.comOra's checks, regroupedEssential 80 points, Recommended 20, bonus capped at 5Free; read-only API, MCP server and CLI, no key
agent-ready.dev72Passing checks over total; bands from 90 to 100 excellent down to 0 to 49 needs improvementFree scan; Pro at 19 USD a month adds API, authenticated MCP and CI gating
SiteSpeakAI scannerNot publishedNot publishedFree web form only
Mintlify Agent Score29, documentation onlySeven categories; per-check weights not foundFree web score and public leaderboard

Compare the standards it checks with the ones you need

List the standards that matter for your site and compare them with the scanner's published list. A useful set spans the stages: robots.txt and bot rules (find), llms.txt, markdown negotiation and JSON-LD (read), and MCP, WebMCP, OpenAPI and payment protocols such as x402 (act).

AgentReady.org, launched on 19 October 2026, lists what a product should implement as 24 identifiers at MUST, SHOULD or MAY. It does not score anyone. By our mapping of check ids, Ora's catalog touches all 24, agent-ready.dev about 15, Cloudflare's scanner about 11 and Mintlify's score about 6. If a vendor does not publish its check list, ask for it in writing.

File checks and agent runs answer different questions

A file check confirms that /llms.txt exists or that a header comes back. A task run sends an agent to do something, such as find a price, and records where it stopped.

Ora's Journey points a real agent at a domain with an intent such as find pricing and records every step, stall and outcome. Ora says every check is reverse-engineered from what moved agent success in its research lab; that is a vendor claim. is-agentic.com shows one observed agent journey in each report as evidence, outside the numeric score. Cloudflare's scanner and agent-ready.dev read files, headers and HTML and do not run an agent on a task.

A grade is only as useful as its published rules

Look for the layers, the points per layer, the number of checks and the grade bands. Ora publishes layer weights and grade bands, but its methodology page and its API catalog disagree on the check count (126 against 132, with Usability at 62 against 68), so record which one you read. agent-ready.dev publishes all 72 check ids, its formula and a dated changelog. Cloudflare lists its checks without weights. SiteSpeakAI lists none.

Record the date of each scan. Rules change: on 2026-10-20 agent-ready.dev stopped crediting an HTML catch-all page as an llms.txt file, and it did not re-score older scans.

Scanners that run from code catch regressions earlier

Readiness slips when pages change, so a check that runs in CI can catch a problem before an agent meets it. Ora documents npx ax audit, a REST API and an MCP server. is-agentic.com offers a CLI, a read-only API and an MCP server without a key. agent-ready.dev puts its API, authenticated MCP server and CI gating in the 19 USD Pro plan. Cloudflare exposes a scan endpoint and an MCP endpoint; the URL Scanner route needs a Cloudflare account and API token. SiteSpeakAI offers a web form only. If your team ships every week, give this question more weight.

The fix path should match the platform you run

Some findings are content fixes and others need infrastructure. Cloudflare's scanner gives a copyable prompt for each failing check. It sits beside Cloudflare features that address several findings directly: Markdown for Agents, AI Crawl Control, Pay per crawl (private beta) and Web Bot Auth verification. agent-ready.dev writes a deterministic fix for each check.

Ora ranks fixes by impact and can re-run selected checks after a fix; its separate fixing product was behind a waitlist on 2026-11-04. SiteSpeakAI's suggested fix is its own paid chatbot product. If you already run on a CDN that offers the fix, the path from finding to fix is shorter.

Check sign-up steps, limits and price before you plan

Cloudflare's scanner is free with no sign-up. Ora's scans are free, anonymous scans are capped at 30 per IP in a rolling 24 hours, and keys that lift the cap are issued by request. is-agentic.com has no paid plans. agent-ready.dev's scan is free and its API is paid. For prices, see what agent readiness tools cost.

Every scanner reads public pages only

Ora does not scan private or logged-in pages, and is-agentic.com does not score authenticated states of a site, so a checkout or account area behind sign-in needs a separate test. We found no scanner among the six that reports sign-ups or revenue; that question belongs to your own analytics.

Ask how quickly checks follow the standards

The standards moved several times in 2026. A2A v1.0 shipped in March, MCP published a stateless specification on 28 July, llms.txt was revised to v2 in August and AgentReady.org reached v2.0 in October. Ask how quickly a scanner updates its checks after a specification changes, and whether it tells you when a score moved because the rules changed rather than your site. Cloudflare's scanner grew from four dimensions at launch to 23 checks in five categories by 4 November. Our news page tracks these changes.

Compare the answers on the same pages

Decide which of the eight questions matter most to your team, weight them, and score each candidate from its public documentation, as the method page does. Then run each shortlisted scanner on the same three pages and one task, and compare the findings rather than the grades. The proof of concept plan lays out the steps.

The Concordance shows what the scanners returned for sixty domains on one day. For two-way comparisons, see Ora and Cloudflare and Ora and is-agentic.com.

Sources

  1. Ora, check catalog. https://ora.ai/api/checks, read 2026-11-04.
  2. Ora, methodology. https://ora.ai/methodology, read 2026-11-04.
  3. Ora, llms-full.txt. https://ora.ai/llms-full.txt, read 2026-11-04.
  4. Cloudflare, isitagentready.com llms.txt. https://isitagentready.com/llms.txt, read 2026-11-04.
  5. Cloudflare, Agent Readiness blog post. https://blog.cloudflare.com/agent-readiness/, read 2026-11-04.
  6. is-agentic.com, methodology. https://is-agentic.com/methodology, read 2026-11-04.
  7. agent-ready.dev, methodology. https://agent-ready.dev/methodology, read 2026-11-04.
  8. agent-ready.dev, comparison with alternatives. https://agent-ready.dev/agent-ready-vs-alternatives, read 2026-11-04.
  9. SiteSpeakAI, AI Agent Readiness Scanner. https://sitespeak.ai/tools/ai-agent-readiness-scanner, read 2026-11-04.
  10. Mintlify, Agent Score launch post. https://mintlify.com/blog/agent-score, read 2026-11-04.
  11. AgentReady.org specification. https://github.com/agentready-org/standard/blob/main/spec/spec.md, read 2026-11-04.
  12. Cloudflare, Markdown for Agents documentation. https://developers.cloudflare.com/fundamentals/reference/markdown-for-agents/, read 2026-09.
  13. Cloudflare, AI Crawl Control documentation. https://developers.cloudflare.com/ai-crawl-control/, read 2026-09.
  14. Cloudflare, Web Bot Auth documentation. https://developers.cloudflare.com/bots/reference/bot-verification/web-bot-auth/, read 2026-09.
  15. A2A releases. https://github.com/a2aproject/A2A/releases, read 2026-09.
  16. MCP, 2026-07-28 release. https://blog.modelcontextprotocol.io/posts/2026-07-28/, read 2026-09.
  17. llms.txt changes. https://llmstxt.org/changes.html, read 2026-09.

Last reviewed 2026-11-05