# Kubernetes Knowledge Compiler

> A compiled, versioned knowledge graph of Kubernetes, built from the official documentation and OpenAPI spec at **compile time**. Every fact is traceable to its source. No runtime LLM is involved in serving this site — it is a static, queryable knowledge resource. The entire site is server-rendered: every node has its own static URL and is readable by any agent that can issue a plain HTTP GET, with no JavaScript required.

Corpus version: v1.34. Nodes: 8198. Relationships: 10754. AI-synthesized documentation: 1210.

## How to use this resource

- The full structured knowledge base is `dataset.json` (nodes + typed edges + provenance). Fetch it and filter by `id` / `type`.
- Every node is also a static HTML page under its section: `/docs/<slug>/` (glossary, pages, concepts), `/api/<slug>/` (API objects + paths), `/rbac/<slug>/` (roles), `/relationships/<slug>/` (control-plane components). Slugs are the node id with `:`/`/`/` `/`{`/`}` remapped to `-`.
- The section index pages list every node as a crawlable link: `/docs/`, `/api/`, `/rbac/`, `/relationships/`, `/learn/`.
- `knowledge.jsonl` is the same data, one node per line. `index.json` is a lightweight title/summary/tag index.
- `sitemap.xml` enumerates every static page so a crawler/agent can discover the full corpus without guessing URLs or running JS.
- `robots.txt` allows all crawlers on all knowledge routes.

## Machine-readable resources

- [dataset.json](https://k8s-docs-compiler.vercel.app/dataset.json) — full knowledge graph (JSON)
- [knowledge.jsonl](https://k8s-docs-compiler.vercel.app/knowledge.jsonl) — same data, one node per line
- [index.json](https://k8s-docs-compiler.vercel.app/index.json) — lightweight title/summary/tag index
- [sitemap.xml](https://k8s-docs-compiler.vercel.app/sitemap.xml) — every static page
- [robots.txt](https://k8s-docs-compiler.vercel.app/robots.txt) — crawler policy (allow all)

## Section index pages (every node linked, server-rendered)

- [/docs/](https://k8s-docs-compiler.vercel.app/docs/) — glossary terms, documentation pages, concepts
- [/api/](https://k8s-docs-compiler.vercel.app/api/) — API objects and API paths
- [/rbac/](https://k8s-docs-compiler.vercel.app/rbac/) — RBAC roles and required-permission edges
- [/relationships/](https://k8s-docs-compiler.vercel.app/relationships/) — ownership chains and control plane
- [/learn/](https://k8s-docs-compiler.vercel.app/learn/) — prerequisite chains (learning paths)
- [/start/](https://k8s-docs-compiler.vercel.app/start/) — guided onboarding path

## Interactive (progressive enhancement, content also in HTML)

- [/](https://k8s-docs-compiler.vercel.app/) — Home — stats + view index
- [/graph](https://k8s-docs-compiler.vercel.app/graph) — Concept Graph — force-directed visualization + server-rendered node/edge table
- [/explore](https://k8s-docs-compiler.vercel.app/explore) — Decision Explorer — kubectl apply flow + RBAC, plus interactive traversal
- [/search](https://k8s-docs-compiler.vercel.app/search) — Search — live full-text box + server-rendered node index

## Note on provenance

Nodes tagged `derived_by` starting with `ai:` were synthesized from source quotes at compile time (confidence < 1.0). All carry `provenance` pointing to the originating document. Prefer source quotes when available.

## Related compiled-knowledge sites (same author, same compile-time-AI approach)

- [Sovereign Knowledge Compiler Explorer](https://skce-explorer.vercel.app/llms.txt) — Sovereign AI concept corpus
- [ChatGPT Compile](https://chatgpt-compile.vercel.app/llms.txt) — 2,489-conversation personal corpus

## Built with

This site follows the compile-time-AI pattern packaged as the [Discoverable Knowledge SDK](https://github.com/kliewerdaniel/discoverable-knowledge-sdk) — compile any dataset into a static, agent-discoverable knowledge site.