What real-world sites teach us about agent readiness.
Practical architectural audits of public web properties: what is observable, what is inferred, and how specific engineering choices make an automated agent journey reliable.
Stripe’s developer surface: why task-shaped documentation is a readiness signal
A public-pattern analysis of how Stripe’s documentation makes API tasks discoverable, typed, and verifiable — without claiming that documentation alone makes a site agent-native.
Read case study →Apple's public web surfaces: clarity is not the same as agent action
A bounded public-pattern case study of Apple's product and accessibility surfaces, separating observable entity clarity from unproven machine actionability.
Read case study →The Verge AI readiness: what a public CMS transition can and cannot prove
A bounded case study of The Verge's documented platform transition and observable publishing surfaces, with a checklist for preserving clarity through a CMS change.
Read case study →Wikipedia AI extraction: why cited, structured pages travel well
A bounded case study of Wikipedia's citation policies, neutral framing, revision model, and machine-readable API, with safeguards against treating retrieval as truth.
Read case study →Reddit UGC and AI: useful context with a hard access boundary
A bounded case study of Reddit's post and comment surfaces, API mechanics, and public-content rules for responsible AI retrieval.
Read case study →Stack Overflow and AI: turning Q&A into attributable context
A bounded case study of Stack Overflow's question-and-answer structure, human validation, attribution, and AI strategy without treating votes as truth.
Read case study →