LOKF Librarian

Maintain .lokf/ - the host repository's knowledge captured as a Linked Open Knowledge Format (LOKF) bundle. LOKF is a semantic profile of OKF: the same directory of Markdown + YAML frontmatter, but every field, type, and relationship is bound to a public vocabulary (schema.org / DCAT / PROV-O), so the bundle expands losslessly to JSON/JSON-LD and RDF and is queryable with SPARQL. This skill covers the full lifecycle: scrape -> build/maintain -> audit -> hand off for review -> keep fresh on a schedule.

Why this matters: plain OKF gives knowledge prose + structure; typical hand-rolled ("vibe-coded") OKF setups add tools that only work in the repo that grew them; LOKF completes the stack - prose + structure + meaning + tools - because binding every field and relation to public vocabularies is precisely what lets standard, schema-generated tooling (JSON Schema, SHACL, SPARQL) validate and query the bundle instead of bespoke scripts.

Scope: this skill owns only .lokf/. A plain, tooling-free sibling okf/ would be owned by some separate okf-librarian type skill. Every LOKF bundle is also a valid OKF bundle - keep the two consistent, but edit each through its own skill. If .lokf/ doesn't exist yet, or the layout below is incomplete, run the lokf-sidecar skill first - it creates knowledge/index.md with the semantic header (Rule 2), knowledge/log.md, the domain directories, plus pyproject.toml and the justfile that just lokf-validate needs. This skill assumes all of that is already in place. Trust verdicts - a person confirming, correcting, retiring, or sending back a concept - belong to the lokf-curator skill: this one hands off to it (section 3) and never writes a human: verification. Readers reach the bundle through lokf-docent, which answers from it and records what it lacked in .lokf/feedback.md for this skill to consume (section 1).