Open Knowledge Format explained

Open Knowledge Format describes knowledge as linked markdown documents. This guide explains the bundle structure and distinguishes portable files from shared semantic meaning.

The Open Knowledge Format is an open specification from Google Cloud that represents a body of knowledge as a directory of plain markdown files. Each concept file carries a small block of YAML frontmatter and a free-form body. The goal is a portable way to hand an AI agent the context it needs, readable by a person and parseable by a machine, with no SDK and no catalog to lock into.

What an OKF bundle contains

A bundle is a folder of markdown files, and the unit inside it is a concept. A concept is anything worth capturing for an agent, such as a table, a dataset, a metric, a runbook or an API. Every concept is one UTF-8 markdown document with two parts, a YAML frontmatter block at the top fenced by a line of three dashes above and below, and a markdown body underneath.

The format asks for exactly one field, type. Everything else is optional, including title, description, resource, tags and a generated block that records who last changed the concept and when. What types exist and what fields most of them carry is left to whoever produces the bundle, though a handful of OKF-defined types, such as Attested Computation, add their own required field on top of the one universal one. Concepts reference each other with ordinary markdown links, so the folder becomes a graph of related knowledge rather than a flat list of files.

Structure versus meaning

Version 0.2 fixes a small set of things and leaves the rest open. It fixes the shape of a bundle as a folder of markdown files, the YAML frontmatter, two reserved filenames and the single required field, all carried forward from version 0.1 unchanged. That is structural interoperability. Any tool can open a bundle and know where the pieces are. Version 0.2 also added optional provenance and trust keys, among them sources, generated, verified, status and stale_after, and it retired two version 0.1 surfaces, since the concept timestamp became generated.at and the body citations list became sources in the frontmatter.

What version 0.2 does not fix is meaning. The format does not say what a metric concept must contain, or how two producers should agree on the same field names. That is semantic interoperability, and version 0.2 leaves it to producers and to conventions that have not been written yet. This is the line to keep in mind when reading announcements about OKF. It standardizes the shape of the files, not yet what the files mean.

Version and status

Google Cloud published version 0.1 in June 2026 and version 0.2 in July 2026. This guide describes version 0.2 as published. A team evaluating OKF should check which version a given bundle or tool targets before comparing it against the description here, because the frontmatter keys have already changed once between 0.1 and 0.2, when the concept timestamp became generated.at and the body citations list became sources. The required field and the two reserved filenames carried forward unchanged.

Where OKF fits with agent-readiness

Agent-readiness, the kind measured by an independent scanner, is about whether an agent can reach and read a public site at all. OKF sits next to that, one layer in. It is a way to package the internal knowledge an agent works from once it is past the front door, the catalog, the metrics and the rules a decision depends on.

So OKF is not a replacement for an llms.txt or a markdown surface on a site. It is the same instinct, plain text an agent can read without a special client, applied to the data and context behind the site rather than the pages in front of it. For a team thinking about what an agent acts on, not only what it can see, that is the part of the picture OKF addresses.

What to do with it today

OKF is new and small, version 0.2 since July 2026, and the semantic half is still open. That makes it worth understanding now and worth watching, but early to build an entire knowledge catalog on. A site that already serves markdown to agents and keeps an llms.txt has the instinct OKF formalizes, and adopting it later will be a short step rather than a rebuild. For an audit of the data behind a site, not only the pages agents see, contact info@turva.dev.

Frequently asked

What is the Open Knowledge Format?

An open specification from Google Cloud that represents a body of knowledge as a directory of plain markdown files. Each concept is one UTF-8 document with a YAML frontmatter block and a body. Google Cloud published version 0.1 in June 2026 and version 0.2 in July 2026.

What does OKF actually standardize?

The shape of the files, not yet their meaning. Version 0.2 fixes the folder of markdown files, the frontmatter, two reserved filenames and one required field. What a concept must contain beyond that field is left to producers at the general conformance level; OKF's own concept types can add their own required fields on top of that floor, for example an Attested Computation concept, where the specification itself makes the runtime field required.

Does OKF replace llms.txt?

No. An llms.txt and a markdown surface make the pages in front of the site readable. OKF packages the knowledge behind it, the catalog, the metrics and the rules a decision depends on. It is the same instinct one layer in.