Learn: lessons and glossary

The nine lessons

• The five areas of agent readiness: find, read, act, pay, trustThe five areas of agent readiness, find, read, act, pay and trust: what each covers, which self-check questions sit in it and the order to work in.
• Why AI agents get stuck on websites: six common wallsSix common walls that stop AI agents on websites, with Ora's dead-end figures: script, refusals, hidden prices, CAPTCHAs, custom sign-in and no owner.
• How to read an agent readiness scanner reportHow to read an agent readiness scanner report: how three scanners group their checks, what a grade is and which failures to act on first.
• robots.txt and bot rules: asking versus enforcingrobots.txt and bot rules in plain terms: asking versus enforcing, the names of AI visitors and what Cloudflare changed in 2026.
• How to read your self-check resultHow to read the self-check result: the five area scores, the level, the cap and what to do with each no and not sure.
• Making pages readable for AI agents: text, structure and markdownMaking pages readable for AI agents: why the first HTML response matters, what to put in it and where structured data and markdown fit.
• Letting AI agents finish the task: forms, sign-in and consentHow to let AI agents finish the task: clear form labels, guest checkout, a standard sign-in and consent at the steps that need the customer.
• How to verify AI agents and bots: a plain-English guideHow to verify AI agents and bots in plain English: why a name proves nothing, what signatures and registration do and what to ask your vendor.
• Agent payments and fraud rules: what leaders need to decideAgent payments and fraud rules in plain language: merchant of record, payment routes, spending limits and a written position for the fraud team.

The glossary

38 terms, each with its own page. Start from the glossary index.