This is the Laurelin Labs Open Knowledge Format (OKF v0.2) bundle: a directory of markdown files with YAML frontmatter carrying type, provenance, trust and freshness metadata, one concept per topic hub, page and entity on this site. It is generated from the same content records that render the visible pages, so it never says more than the site does. Honest status: OKF is a versioned Google Cloud specification with a real trust model, but it defines no web discovery mechanism and no search or answer engine fetches site bundles today. We publish it as zero-cost future-proofing for the agentic web, not as a ranking factor, and we do not publish llms.txt.

How does an agent use it?

Start at index.md (progressive disclosure), follow bundle-relative links, and read each concept's frontmatter before its body: generated says who wrote it and when, verified(absent until a person has reviewed) sets the trust tier, and stale_after is the instant after which the concept should be re-fetched. Footnotes join to sources[].id. The bundle is also declared in every page's head as link rel="alternate" type="text/markdown".

Which files are in the bundle?

Read the reasoning in guardrails, agentic and future-proofing.

AI search visibility knowledge bundle for AI agents

How does an agent use it?

Which files are in the bundle?