standard
Markdown convention introduced in 2024, spec hosted at llmstxt.org, placing a curated reading list for AI assistants at /llms.txt.
No evidence-backed relationships are recorded.
FoxyInvoice prerenders every public Angular route to static HTML at build time so crawlers that skip JavaScript get the full page text. A build step and a one-line drift check take the place of the always-on SSR server its team declined to run.
Reality
Version 2 of the llms.txt proposal landed in August, Lighthouse has audited the file since Chrome 13.3, and three of the four big WordPress SEO suites generate one. Google Search still says nothing fetches it.
Reality
The llms.txt spec picked up link relations and subpath scope in a quiet 10 August revision. The fetch logs suggest the file gets read only when a client has already been told to look for it.
Reality
The CLI does generative engine optimization, but GEO means geographic across most of the indexed web, so a model given only the name picked the wrong sense and then supplied a vendor to fit it. Closing the gap took three files.
Reality
Jeremy Howard's convention has been public since September 2024, and no large model provider has confirmed reading the file, so the case for spending an hour on one rests on what the drafting forces you to decide.
Reality
Suman Debnath ran a week-long answer-engine pass, watched ChatGPT name him two days later, and then spent most of the write-up arguing that the sequence is not a cause. The cross-device check and the three silent assistants are why that argument holds.
Reality
A four-layer model of how coding agents choose an SDK pushes the work past documentation into llms.txt and MCP servers. The search layer comes with a measured trigger rate from Vercel; the layers below it come with none.
Reality
A dev.to walkthrough counts roughly 303,800 extensions in the Chrome Web Store and 49,518 new ones in a single quarter. Agents can draft an entire listing in an afternoon, but the store still checks it against your code.
Reality
PostHog put 344 visitors against exactly 344 pageviews on one country row, which is arithmetic no crowd of humans commits, and whatever produced it ran enough JavaScript to sail past a server-side filter.
Reality
A scan of 6,214 corporate domains found 120 sites whose AI-facing docs point at code packages and domains nobody owns, and registering a few of those names produced installs traced back to Claude, Codex and Hermes agents inside real networks.
Reality
Docs7 takes a Mintlify config, serves Markdown from the URLs agents already request, and counts which AI products fetch what, a metric Context7 itself says measures provider visits rather than readers.
Reality
The file exists so an agent can skip your documentation and read the install steps instead. Pandex checked 8,565 of them, found 237 references pointing at names anyone could register, and registered one.
Reality
A CDN rule aimed at AI crawlers matched the Vendor/Language User-Agent shape that the official OpenAI and Anthropic SDKs send by default. The harness meant to catch it sent an allowlisted string of its own and reported green.
Reality
Ownware's own access log records an AI coding assistant fetching /legal, /support and security.txt within minutes, then guessing /contact and stopping at the first 404, fifteen hours before the human paid.
Reality
The Hanover Institute has pushed out more than 100 unbylined, AI-written articles in under a month: question headlines, citations that do not link, and a crawler map for the models reading it.
Reality
A 1,180-record knowledge base serves crawlers and agents from the same 0.39-second build. The MCP endpoint imports what the build wrote, because two derivations of one join eventually disagree.
Reality
A SparkToro estimate puts clickless US Google searches at 68.01% for the first four months of 2026. The counters that survive that are citation and what a visitor does after the answer.
Reality
Promptway ran the public is-agentic grader against itself and lost most of its points to missing contact pages, a dead-end 404 and broken content negotiation. The rest it lost on purpose.
Reality
The experimental category grades semantics, CLS, llms.txt and WebMCP, and it is explicitly not a ranking factor. The scoring rule means most sites can pass it without touching an agent.
Reality
One developer measured 7,032 plugins in the WordPress.org directory. Phase one of search is a hard keyword filter; phase two is ranked partly on installs a new plugin does not have.
Reality
Build · August 15, 2026 · 1 publisher