// insights

How to be read by machines

One guide per check our readiness tool runs. What it means, why it matters, and exactly how to fix it.

DiscoveryStructured dataRenderingProtocolsCommerce

What is llms.txt, and does your site need one?

llms.txt is a markdown map of your site for AI assistants. Here's what goes in it, where it lives, and how it differs from robots.txt and sitemap.xml.

Read

Discovery6 min

Which AI crawlers is your robots.txt blocking?

GPTBot, ClaudeBot, PerplexityBot and Google-Extended each read robots.txt. A single Disallow can make your site invisible to AI answers — often unintentionally.

Read

Structured data7 min

JSON-LD is how AI answer engines know what you are

Structured data turns a page into facts an engine can quote with confidence. Which schema types matter, and the @graph mistake that makes valid markup look invalid.

Read

Rendering6 min

Most AI crawlers don't run your JavaScript

If your content only appears after hydration, many AI crawlers see an empty page. How to check what they actually get, without guessing.

Read

Protocols8 min

Turning an OpenAPI spec into an MCP server

If you already publish an OpenAPI spec, most of an MCP server is derivable from it. What maps cleanly, what doesn't, and where generation stops helping.

Read

Protocols4 min

The .well-known/mcp.json manifest, explained

A manifest is how an AI client discovers that your product is operable, not just readable. What belongs in it and when publishing one is premature.

Read

Commerce9 min

UCP, ACP and AP2: the agentic commerce stack

Three protocols, three different jobs — discovery, checkout and payment authority. What each does, who's behind them, and what to implement first.

Read

Commerce5 min

Product markup thin enough to get skipped

Shopping agents compare options across sites and drop the listings they can't fully parse. The Product and Offer fields that decide whether you're comparable.

Read

Structured data5 min

Canonical URLs, titles and descriptions in an AI-answer world

The oldest SEO tags still matter, for a new reason: they decide which URL gets cited and what an engine says about you when it does.

Read

Discovery4 min

Sitemaps still matter, for a boring reason

A sitemap is how a crawler learns your pages exist without following every link. Cheap to generate, and the input for a decent llms.txt.

Read

Structured data3 min

OpenGraph tags: the fallback summary you already own

og:title, og:description and og:image aren't just for social previews — they're a clean, duplicated summary of the page that anything can read.

Read

Want to know which of these apply to your site? Run the free checker.

Check your site

// ready_when_you_are

Let's build your agentic advantage

Tell us what should be automated, connected or sold. We'll map the fastest path from idea to a production system.

Start a project Explore services