Faro Editorial
Practical guides on AI visibility, structured data, llms.txt, and how to make sure AI agents find, understand, and recommend your business.
How to build an MCP server in five steps: choose the actions agents should take, generate a starter or write it with the official 2026 SDK, test it in the MCP Inspector, host it on public HTTPS for Claude and ChatGPT, and get it found. Only 12.0% of 13,164 sites in the Faro Score Index pass the MCP server check.
A GEO audit checks whether AI assistants can reach your site, understand what you sell, trust who you are and name you to buyers. Here are the five checks, a 30-minute method you can run free, and what 13,145 scanned sites get wrong.
We compared 1,803 WordPress sites in the Faro Score Index with 1,642 on other platforms. Overall readiness is identical, one in four WordPress sites opts out of GPTBot training, and 11% refuse GPTBot even though robots.txt allows it.
Core Web Vitals don't predict AI search visibility. Here's what actually determines whether an AI crawler's fetch succeeds, and how to test your own server response time in under a minute.
A plain explanation of the Agentic Commerce Protocol (ACP), the open standard from Stripe, OpenAI, and Meta that lets AI agents browse, check out, and subscribe on a buyer's behalf, and what your pricing and product pages need to expose before an agent can act on them.
A buyer's checklist for evaluating any AI visibility tool, including Faro: nine checkable criteria covering crawler access, structured data, citation tracking, and competitive benchmarking.
Google added a new user-triggered fetcher, Google-Agent, behind Gemini's auto browse feature in Chrome. Here is what it does, how Google verifies it, and whether you should block it.
A traced study of 1,033 agent runs found sitemaps fetched directly in only 4% of runs. Here is what ChatGPT, Claude, and Gemini actually fetch instead, and what still helps them find your pages.
Original Faro data on AI crawler readiness across 815 websites as of 16 October 2026: how many hide content behind JavaScript, return soft 404s, leave llms.txt unlinked, or block AI crawlers, and what to fix first.
36 articles published