How to read an agent readiness scanner report
In short
A scanner checks a list of files, settings and page features and turns the result into a grade or a pass count. Each tool groups its checks differently, so scores from different tools are not comparable. Read the failed checks that sit on your main task, and treat the grade as a band rather than a verdict.
Three scanners, three groupings
- Read 4 Nov 2026
Ora scores four layers (Discovery, Access, Usability, Payments) normalised to 100, with grades from A+ to F.
- Read 4 Nov 2026
The scanner checks five categories: discoverability, content accessibility, bot access control, protocol discovery and commerce (x402, MPP, UCP, ACP).
- Read 4 Nov 2026
is-agentic.com weights essential checks most and excludes checks that do not apply, so a content site is not penalised for having no API.
What to read
- The grade or score as a band, not a number.
- The failed checks, sorted by the page they sit on. A failure on your checkout matters more than one on an old blog post.
- The checks the tool left out as not applicable. Not applicable is not the same as passed.
A scanner reads public pages. It cannot see a checkout or account area behind a login, and it cannot tell you whether an agent finished a task.
What to do next
Sort each failure into a decision or a technical fix, give it an owner and a date, and re-run the scan when the fix is in. The guide on reading a scanner result without a developer walks through it.
Related
- Tools that checkThree free scanners.
- Reading a scanner resultFive steps.
Sources
- Ora llms.txt (ora.ai/llms.txt), read 4 Nov 2026. Vendor claim.
- Is Your Site Agent-Ready? (Cloudflare, isitagentready.com) (isitagentready.com), read 4 Nov 2026.
- Is Agentic: AI Agent Readiness Score for your Site and App (is-agentic.com), read 4 Nov 2026.
Last reviewed