agent-readiness-audit
Agent Readiness Audit
Score whether an AI agent acting for a buyer can find you, evaluate you against alternatives, and do something. This is a different question from whether an AI engine cites you in an answer, and the two come apart more often than people expect: a company can be well-cited in AI answers and still be impossible for an agent to transact with, because its pricing is a contact form and its content only exists after JavaScript runs.
The shift this anticipates is real but not yet fully arrived. Gartner projects that by 2028 a large majority of B2B buying will be intermediated by AI agents. Whether that specific figure lands, the direction is visible now: buyers already use AI to build shortlists, and a company an agent can't parse is a company that doesn't make the list.
The awkward part is that failure is silent. You don't get a bounce, a failed form, or a bad review. You're simply absent from a comparison you never knew happened.
When to use
The user is asking whether AI agents can buy from, evaluate, or act on their business — or is planning for agentic commerce, MCP exposure, or machine-readable product data. For whether AI engines cite their content, use seo-geo-aeo-audit or geo-content-optimization instead.
Before you start
- Read the brand/product context (
.agents/product-marketing.md) for what's sold and how. - Check rendering yourself. View-source a key page — not DevTools, which shows the post-JavaScript DOM. Search the raw HTML for a sentence you wrote. This one check decides most of the score.
- Read robots.txt. Note which AI crawlers are allowed and which are blocked, and whether that was a decision or a default.
- Try to buy your own product as a machine would — with no login, no human, and only what's on the public site.