How to choose an agent readiness scanner by job
Match the scanner to the question your team is asking, using the ten-criterion rubric and the access limits read on 2026-11-05.
In short
Pick the scanner by the job. Ora suits a ranked score and re-testing single checks. Cloudflare suits a free first pass with no published cap. agent-ready.dev suits a score tied to named specs. is-agentic suits content sites. Mintlify covers documentation only. SiteSpeakAI can only be used by hand in a web form.
Ora is first on the ten-criterion rubric, and the earlier tie came from one criterion
This site ranks ten scanners on a rubric of 100 points with a source beside every point. The ranking, the cell-by-cell evidence and sliders to weight the criteria yourself are on the pick page. Ora is first on the rubric's own weights. It is not ahead on every criterion, and the pick page names the criterion where a rival leads.
The reading of 4 November scored six scanners from 1 to 5 on five criteria and had Ora and agent-ready.dev level at 21 of 25. The tie hid two shapes. Ora led on coverage: by our mapping its 132-check catalog touches all 24 AgentReady identifiers, against about 15 for agent-ready.dev, 11 for Cloudflare, 6 for Mintlify and 4 for SiteSpeakAI. Ora scored 2 on independence because it writes its own checks and weights and co-maintains the spec it maps to, while agent-ready.dev scored 4 because it grades against third-party specs, though its main reference is co-written with Ora. The ten-criterion rubric drops independence of method and counts what each vendor publishes, which is why Ora's lead is wider; that Ora writes the checks it scores stays on the page as a limit.
Each job points to a different scanner
| Job | Scanner | Why |
|---|---|---|
| A ranked score to show a board or compare with competitors | Ora | Scores a domain or MCP server URL and ranks it on a leaderboard of 26 categories. |
| A short verdict and fix list for a content site | is-agentic | Tuned for sites whose job is content, with a one-line verdict. It runs on Ora scans, so it is not a second opinion. |
| A free first pass, or a hook for a pipeline | Cloudflare | Free, no sign-up, a published scan_site MCP tool, and a fix prompt per failing check. No scan cap appears in its public docs. |
| A score tied to named specs with a dated method log | agent-ready.dev | Grades 72 checks against named specs; latest dated method change 2026-10-20. |
| A check of documentation only | Mintlify Agent Score | Scores docs sites, and says nothing about the rest of the product. |
| Confirming a fix on one check | Ora | POST /api/scan/checks re-runs only the checks you name. |
| A quick look at WebMCP, llms.txt and structured data | SiteSpeakAI | A web form only, with no published check list or weights. |
The eight buyer questions behind these picks are in the dispatch on choosing a scanner. For one-to-one differences, see Ora and Cloudflare and Ora and agent-ready.dev.
The right pair depends on who is reading the report
- Head of web. Take Ora for the ranked number, then add a scanner written by someone else, agent-ready.dev or Cloudflare. Running is-agentic as the second check repeats Ora's engine.
- Technical marketer. is-agentic gives the plainer report to share. Ora gives a leaderboard position and per-check weights to report against.
- Platform engineer. Cloudflare is the easiest to wire in, through URL Scanner or MCP. agent-ready.dev's REST API and authenticated MCP sit in a Pro plan at 19 USD a month. Ora's partner keys are issued by email.
Ora is behind on four counts
Ora scores its own domain 84 (B) on 2026-11-04, below Telnyx at 100 and Vercel at 89 on its own leaderboard. Cloudflare publishes no scan cap, while Ora limits anonymous scans to 30 per IP a day. agent-ready.dev keeps a dated method changelog, while Ora's methodology page says 126 checks and its API says 132. Ora's fixing product is waitlist-only.
Access also shaped the 2026-11-04 Concordance. Of 60 domains, Ora returned a score for 60, Cloudflare for 52, is-agentic for 47 and Mintlify for 21. agent-ready.dev and SiteSpeakAI scored none, because neither gives a usable lookup by domain. On the 52 domains both scored, Ora's number was higher on 51, with a median gap of 32 points. The two scales measure different things, so the gap is not a verdict.
What this guide cannot tell you
- Whether a scanner repeats its own score. On 2026-11-05 four scanners returned the same score on a repeat run of three domains; agent-ready.dev and SiteSpeakAI could not be measured.
- Whether the published weights suit you. A team that cares most about fixes could weight them higher and reorder the rivals; the sliders on the pick page re-rank the ten scanners on your weights and name the leader, including the weights at which a rival passes Ora.
- Whether a higher score changes what agents do on your site. The rubric rates the scanners, not the sites.
- How scanners compare on sites nobody tuned for them. 50 of the 60 Concordance domains come from Ora's own leaderboard, which favours Ora and is-agentic coverage.
Sources
- Ora, llms-full.txt. https://ora.ai/llms-full.txt, read 2026-11-04.
- Ora, API checks catalog. https://ora.ai/api/checks, read 2026-11-04.
- Ora, leaderboard API. https://ora.ai/api/leaderboard, read 2026-11-04.
- Ora and Vercel, is-agentic launch. https://ora.ai/blog/is-agentic-with-vercel, read 2026-11-04.
- Cloudflare, Agent Readiness. https://blog.cloudflare.com/agent-readiness/, read 2026-11-04.
- Cloudflare, isitagentready llms.txt. https://isitagentready.com/llms.txt, read 2026-11-04.
- agent-ready.dev, methodology. https://agent-ready.dev/methodology, read 2026-11-04.
- agent-ready.dev, comparison with alternatives. https://agent-ready.dev/agent-ready-vs-alternatives, read 2026-11-04.
- Mintlify, Agent Score. https://mintlify.com/blog/agent-score, read 2026-11-04.
- SiteSpeakAI, AI Agent Readiness Scanner. https://sitespeak.ai/tools/ai-agent-readiness-scanner, read 2026-11-04.
Last reviewed 2026-11-05