okf

name	okf
description	Create, validate, and consume Google's Open Knowledge Format (OKF) bundles — YAML-frontmatter Markdown files with type / title / description / resource / tags / timestamp fields for portable, interoperable AI-agent knowledge sharing. OKF formalizes the LLM-Wiki pattern into a vendor-neutral open specification so any producer can write and any agent can consume without translation. Routes: use `llm-wiki` for raw source capture + vault maintenance, `obsidian` for Obsidian-vault workflows, `graphify` for durable committed graphs, `scrapling` for web-content extraction into OKF docs. Triggers on: okf, open knowledge format, knowledge bundle, okf document, llm wiki standard, knowledge atom, agent context format, okf frontmatter, okf bundle, knowledge interoperability.
allowed-tools	Read Write Edit Bash Grep Glob WebFetch
compatibility	Universal — pure Markdown + YAML, no runtime dependencies. Works in any agent that can read and write files (Claude Code, Codex, Gemini CLI, OpenCode, Cursor, Copilot). Plugin-installable via `npx skills add`. Routes raw source collection to `llm-wiki`, Obsidian vault automation to `obsidian`, durable knowledge graphs to `graphify`, and web scraping into OKF docs to `scrapling`.
metadata	{"tags":"okf, open-knowledge-format, knowledge-bundle, llm-wiki, knowledge-graph, metadata, context-engineering, interoperability, yaml-frontmatter, google, agent-context","platforms":"Claude, Codex, Gemini, Cursor, Copilot, OpenCode, All","keyword":"okf","version":"1.0.0","upstream":"https://github.com/google/open-knowledge-format","source":"akillness/jeo-skills","license":"MIT"}

OKF — Open Knowledge Format

Open Knowledge Format (OKF) is an open specification proposed by the Google Cloud Data team that formalizes the LLM-Wiki pattern into a portable, interoperable format. Knowledge is stored as YAML-frontmatter Markdown files inside a directory (a "bundle"). Any producer can write it, any agent can consume it — no proprietary SDK, no special runtime, no lock-in.

Core idea: Just markdown. Just files. Just YAML frontmatter.

When to use this skill

• You need to create structured knowledge documents an AI agent can parse — table schemas, API specs, metric definitions, runbooks, deprecation notices
• You want a portable knowledge bundle that survives tool changes (no vendor catalog dependency)
• You're building a multi-agent context layer where different systems need the same organizational knowledge (join paths, business KPIs, incident guides)
• You want to cross-link knowledge atoms into a richer relationship graph than a parent/child hierarchy allows
• You need to publish or distribute knowledge as a tarball, git repo, or filesystem mount
• You want to formally align your LLM-wiki output with the OKF v0.1 standard

When NOT to use this skill

• Raw web-source capture and vault-level wiki maintenance → llm-wiki
• Obsidian vault automation (plugin dev, CLI, URI handoff) → obsidian
• Durable committed knowledge graphs from code/docs → graphify
• Web scraping to feed knowledge documents → scrapling

OKF Structure

The bundle directory

knowledge-bundle/
├── tables/
│   ├── orders.md
│   └── customers.md
├── metrics/
│   └── wau.md
├── runbooks/
│   └── outage-response.md
└── concepts/
    └── revenue-calculation.md

Each file is a knowledge atom — one YAML-frontmatter Markdown document. The directory itself is the distributable unit (tarball / git / filesystem mount).

Frontmatter schema (OKF v0.1)

---
type: <string>          # (required) kind of atom: "BigQuery Table", "API Endpoint",
                        #   "Concept", "Runbook", "Metric Definition", "Data Source", …
          
    
          
 [, ]