University
In short
Nine short lessons in three tracks take a reader from what agent readiness means to running tests and keeping a score from slipping. Each lesson names the standards it relies on, cites its sources and says what it cannot tell you. The glossary defines the forty terms the lessons use.
Track 1. Basics
- Lesson 1. The five layers of agent readinessOne question per layer, and the files or protocols that answer it.
- Lesson 2. How AI agents read a web page: HTML, markdown and llms.txtWhat arrives in the response, and three checks to run first.
- Lesson 3. How websites recognise AI agents: user agents, robots.txt and signed requestsThree signals, and which of them is verification.
Track 2. Choosing tools
- Lesson 4. Testing tools and fixing tools: the two kinds of agent readiness toolWhy a scanner and an edge feature are not substitutes.
- Lesson 5. How to read an agent readiness scoreWhat goes into the number, and what it cannot tell you.
- Lesson 6. File checks and task runs: two ways to test agent readinessOne tells you the signals are present. The other tells you whether an agent got through.
Track 3. Running it
- Lesson 7. How to run your first agent task testOne task, four page checks, and a written record of where the agent stopped.
- Lesson 8. Keeping agent readiness from regressingReadiness is a property of this week's release, not a project you finish.
- Lesson 9. Setting a policy for AI crawlers and agentsTraining, search and assistants are three decisions, not one.
Reference
- GlossaryForty terms with sourced examples.
- FAQ
- Standards Map
Sources
This page cites no outside source of its own. It is built from the pages it links to, each of which lists its sources, and from the rubric on the method page.
Last reviewed 2026-11-05