When There's No API, the Browser Is the API
An AI agent is only as capable as the systems it can reach. Plenty of the platforms a business actually runs on expose no API, or gate it behind an enterprise tier. But every one of them has an interface a human uses every day, and anything a human can do in a browser, an agent can do too.
Read →Serving Markdown to Bots with Content Negotiation
A page that costs you a full server-side render to produce can be served to a crawler as a few kilobytes of Markdown that cost almost nothing. Content negotiation lets you do exactly that: same content, cheaper representation, and far friendlier rate limits for the bots that matter.
Read →GEO vs SEO: Where the Boundary Actually Is
The honest answer is neither 'GEO is a new discipline' nor 'GEO is just SEO with AI readability'. GEO and SEO share the retrieval and authority substrate but diverge at the objective function and the content layer.
Read →llms.txt and llms-full.txt: A Sitemap, Rewritten for Language Models
sitemap.xml answers a search crawler's question, which URLs exist and when did they change. It answers none of a language model's: what is this site about, what is worth reading, and can I have the content without rendering the pages. llms.txt and llms-full.txt are the answer that has stuck.
Read →The Call Before the Transfer
When a caller asks to be put through to a specific person, the easy thing is to dial that person and connect the two. The polite thing is to ask first. Doing that without a human operator in the middle, while the caller waits live on the line, turned out to be mostly a problem about time.
Read →