Open Knowledge Format (OKF v0.2) Adapter
This adapter provides full Open Knowledge Format (OKF) v0.2 interoperability for Copilot LLM Wiki, allowing your personal markdown second brain to be exported into a standardized, machine-readable knowledge bundle for enterprise AI systems, Google Cloud agents, and multi-agent orchestrators.
Architecture: Clean Core + Adapter Pattern
Rather than forcing mandatory YAML frontmatter and multiple sub-indexes onto your daily desktop note-taking workflow, this repository separates concerns:
- Core Wiki (
wiki/): Maintained as clean, fluid, standard Markdown. Optimal for rapid human note-taking and distraction-free viewing in Obsidian, VS Code, and JetBrains. - OKF Adapter (
adapters/okf/): Transforms the clean wiki into an official OKF v0.2 bundle on demand, deterministically generating YAML frontmatter, actor conventions (copilot-librarian/1.0), timestamps, and progressive disclosure sub-indexes.
🚀 Exporting to an OKF v0.2 Bundle
Run the Python exporter from the repository root:
python adapters/okf/export.py
The script will compile your wiki into dist/okf/:
- Injects standard OKF v0.2 YAML frontmatter (
type,title,description,status,generated). - Creates
index.mdfiles in each subdirectory (entities/,concepts/, etc.) with asterisk bullet syntax. - Generates a root
dist/okf/index.mdwithokf_version: "0.2"frontmatter. - Formats
dist/okf/log.mdinto newest-first dates.
🤖 Optional: Native OKF Copilot Persona
If you want your desktop Copilot instance to author notes directly with OKF frontmatter during interactive chat:
- Replace or supplement your
.github/copilot-instructions.mdwithadapters/okf/copilot-instructions.okf.md. - Copilot will write OKF frontmatter natively on every newly generated concept or entity note.
📜 Specification Reference
For complete schema details, trust tiers, and lifecycle definitions, consult SPEC.md.
Original OKF v0.2 specification structure contributed by @aayush-t-gilead.