Google published a format for turning a body of knowledge into a folder of linked markdown files. It was built for internal company data, and by accident, it solves a problem public websites have too. Right now, the most an AI agent gets from your website is a flat read of your pages, one at a time. This format builds a graph of how your ideas connect instead, so I tried it on my own website.
Google’s Open Knowledge Format Is A Directory Of Linked Markdown Files
On June 13, 2026, Google’s data team published the Open Knowledge Format, or OKF, a way to represent a body of knowledge as a directory of markdown files with a thin layer of YAML frontmatter. Each concept, a table, a metric, a runbook, an API, gets its own markdown document. A short block of YAML carries the queryable fields, type, title, description, resource, tags, and timestamp; the markdown body carries the explanation, and concepts link to each other with ordinary markdown links, which Google says turns the directory into “a graph of relationships.” There is no runtime, no SDK, no build step. Google describes a bundle in three phrases: “just markdown,” “just files,” “just YAML frontmatter.”
The target is internal company knowledge, the context Google says is “locked behind whichever surface created it,” and it is early, v0.1, which Google calls “a starting point, not a finished standard.” Nothing in the announcement mentions public websites. That gap is what this piece is about.
On A Website, A Knowledge Graph Beats A Flat Page-Copy
The agent-readable version of your website, the one a model or a browser actually consumes, is flat. Serving each page as markdown, the way Cloudflare does at the network edge, is close to AMP for LLMs: a second, stripped copy of every page for a machine to read. It mirrors what you already have, page for page, and it drops the same thing every page-by-page copy drops, which is how the pages relate to each other.
A knowledge graph keeps that relationship layer. When your concepts link to each other, an agent not only learns what each one is, but it also learns how they sit relative to each other, which is most of what understanding a website actually means. Two pages can both mention a concept and never tell a machine that one is the framework underneath it and the other is the narrower goal beside it. A graph says it outright, in links the machine follows. OKF is an off-the-shelf way to build that graph: markdown, so it is cheap, and structured, so it carries the relations.
I Tried OKF On The No Hacks Website
I wrote an OKF bundle for the No Hacks website, one markdown file each for the brand, the host, Machine-First Architecture, the agentic web, Agent Experience Optimization, Answer Engine Optimization, llms.txt, and WebMCP. Each follows Google’s conventions, the YAML fields on top and a plain markdown body underneath. The work was mostly deciding which concepts mattered and how they connect, not writing the…
Source link
Disclaimer
We strive to uphold the highest ethical standards in all of our reporting and coverage. We blogs.grocliq.com want to be transparent with our readers about any potential conflicts of interest that may arise in our work. It’s possible that some of the investors we feature may have connections to other businesses, including competitors or companies we write about. However, we want to assure our readers that this will not have any impact on the integrity or impartiality of our reporting. We are committed to delivering accurate, unbiased news and information to our audience, and we will continue to uphold our ethics and principles in all of our work. Thank you for your trust and support.
Website Upgradation is going on for any glitch kindly connect at [email protected]
