跳到正文

Albertchamberlain

Awesome-OKF

Curated Open Knowledge Format (OKF) resources — plugins, skills, proposals, and tools. YAML-driven, agent-searchable.

README 已保存到本站,可直接阅读

Documentation snapshot

README 快照

这篇是英文原文

下面正文是项目自己的英文 README。想读全文就用浏览器自带的整页翻译: Chrome / Edge 点地址栏右侧的翻译图标,或用右键菜单里的「翻译成中文」; 手机浏览器一般在菜单里。

本页保存的是公开项目资料快照,阅读过程不需要连接 GitHub。

Awesome OKF

The curated catalog of Open Knowledge Format resources.

YAML-driven. Agent-searchable. Community-curated.

English | 中文 | 日本語 | 한국어

Catalog  ·  Connect an Agent  ·  CLI  ·  Contributing


What is OKF

Open Knowledge Format (OKF) is an open specification by Google Cloud — define knowledge as a directory of Markdown files with YAML frontmatter and a small set of conventions. No runtime, no SDK.

What’s Different Here

Awesome OKF keeps the entire OKF ecosystem in one validated YAML catalog, then turns it into a browsable list, a searchable CLI, and an MCP meta-server that AI agents can query directly:

                      catalog.yaml
                           │
          ┌────────────────┼────────────────┐
          ▼                ▼                 ▼
      README.md         CLI tools        MCP server
   (human-browsable)  (searchable)   (agent-searchable)

Edit one record. Regenerate the docs. Re-query from anywhere.


See It in Action

User (or Agent):
  "Find an OKF plugin for Obsidian vault conversion."

Agent calls:
  search_catalog({
    "query": "Obsidian",
    "kind": "plugin",
    "limit": 3
  })

Awesome-OKF responds:
  ┌──────────────────────────────────────────────────────────────┐
  │ obsidian-to-okf                                plugin        │
  │ Convert Obsidian vaults to OKF — wikilinks become OKF links. │
  │ Platform: python  ·  Tags: obsidian, wikilink, markdown      │
  └──────────────────────────────────────────────────────────────┘

The catalog speaks OKF.


Quick Start

Connect to Your Agent

Add Awesome OKF to any MCP client so your agent can discover OKF resources:

pipx install awesome-okf
{
  "mcpServers": {
    "awesome-okf": {
      "command": "awesome-okf-server"
    }
  }
}

CLI

awesome-okf stats
awesome-okf list --kind plugin
awesome-okf search obsidian
awesome-okf readme

🛠️ Our Tools

convert-to-okf 🔄

Zero-dependency CLI that converts various formats into OKF knowledge bundles:

📥 Input Format✨ What It Does
📋 Markdown awesome-xx listsExtracts - Title — Description items → OKF entries
📊 JSON arraysConverts {title, url, description} objects → OKF entries
🔗 URL listsPlain text URL collections → OKF entries
# One command, instant OKF bundle
python scripts/convert-to-okf.py README.md -o kb/ -t concept

# Output: kb/ with 45 Markdown files, each with YAML frontmatter
# Ready for: myokf validate kb/

awesome-okf CLI 🎛️

The data-driven catalog CLI (same architecture as Awesome-MCP):

🔍 Command📝 Purpose
awesome-okf search obsidianFull-text search across all 29 entries
awesome-okf list --kind pluginFilter by category
awesome-okf readmeRegenerate this README from catalog.yaml
awesome-okf-serverMCP meta-server — let AI agents query the catalog

🏆 Repository📌 What It Offers
⭐ yzfly/awesome-okf中文世界第一个 OKF 落点 — 7 plugins + 7 skills + 3 proposals
⭐ linyiru/awesome-okfEnglish OKF resource hub — spec, tools, samples, guides
📚 GoogleCloudPlatform/knowledge-catalogOfficial OKF spec, SDK, and proposals by Google
🧠 karpathy/llm-wikiThe original LLM Wiki that inspired OKF

Catalog

29 curated entries · 4 tools · 7 plugins · 7 skills · 5 proposals · 6 docs Deliberately curated — not an exhaustive index.

🛠️ Tools & CLI

Cli

  • myokf-cli cli — Unified CLI for OKF — pull from GitHub, validate, and package to single-file web. — cli, python, validation, packaging

Conversion

  • convert-to-okf cli — CLI tool to convert Markdown awesome-xx lists, JSON arrays, and URL lists into OKF knowledge bundles — zero dependencies, standard library only. — conversion, cli, markdown, json

Quality

  • OKF Validator (myokf) cli — Built-in OKF schema validator — checks YAML frontmatter, link integrity, and spec compliance. — validation, quality, schema

SDK

  • OKF Python SDK ✅ python — Official Python SDK for reading, validating, and writing OKF bundles — referenced by Google as the reference implementation. — sdk, python, official

🔌 Producer Plugins

Cli

  • myokf-cli (plugin entry) python — Unified CLI entry point wrapping all seven producer plugins. — cli, aggregator, zero-dependency

Code

  • github-to-okf python — Extract code symbols from GitHub repositories into OKF. — github, code, symbols, zero-dependency

Document

  • feishu-to-okf python — Convert Feishu (Lark) knowledge spaces and documents into OKF. — feishu, lark, document, zero-dependency
  • notion-to-okf python — Convert Notion Markdown exports into OKF. — notion, markdown, zero-dependency
  • obsidian-to-okf python — Convert Obsidian vaults to OKF — wikilinks become OKF links. — obsidian, wikilink, markdown, zero-dependency

List

  • awesome-to-okf python — Convert GitHub awesome-xx lists into structured OKF knowledge bases. — awesome-list, conversion, zero-dependency

Web

  • html-to-okf python — Convert HTML files into OKF. — html, web, zero-dependency

🤖 Claude Code Skills

Conversion

  • book-to-okf claude-code — Split books and long-form articles into interlinked concept knowledge bases. — book, long-form, concepts
  • code-to-okf claude-code — Convert codebases into OKF with Claude Code. — code, repository, enrichment

Creation

  • okf-creator claude-code — Create high-quality OKF knowledge bases from scratch with Claude Code. — creation, knowledge-base

Import

  • awesome-to-okf (skill) claude-code — Import awesome lists and enrich them into OKF with Claude Code. — awesome-list, import, enrichment
  • github-to-okf (skill) claude-code — Repository to OKF enrichment workflow with Claude Code. — github, repository, enrichment

Publishing

  • okf-to-book claude-code — Publish OKF knowledge bases as VitePress documentation sites. — vitepress, publishing, docs
  • okf-to-web claude-code — Package OKF into a single-file web page with interactive knowledge graph. — web, single-file, knowledge-graph

📝 Proposals & Extensions

Upstream

  • Attested Computation Proposal web — Propose verifiable computation records for OKF knowledge entries. — computation, verification, upstream
  • Lifecycle & Staleness Proposal web — Propose status and stale_after lifecycle fields for OKF v0.2. — lifecycle, staleness, upstream
  • OKF Discovery Protocol (KEP-1) ✅ web — Proposal for a standard discovery mechanism that lets agents find OKF bundles without hardcoding paths. — discovery, upstream, kep
  • Sources & Provenance Extension web — Propose sources field and provenance tracking for OKF v0.2. — sources, provenance, upstream
  • Trust Signals (KEP-2) ✅ web — Proposal for verification chains and trust tiering so consumers can distinguish machine-confirmed from human-reviewed content. — trust, verification, upstream, kep

📖 Documentation & Specifications

Example

  • Karpathy’s LLM Wiki (OKF) web — Karpathy’s LLM knowledge base converted to OKF — a real-world example of OKF in action. — example, llm, karpathy
  • LLM Wiki (Karpathy) web — The original LLM knowledge base by Andrej Karpathy that inspired OKF — a living wiki of LLM concepts as Markdown files. — example, llm, karpathy, inspiration
  • OKF Market Concept web — A conceptual OKF knowledge market — imagine a marketplace where knowledge entries are traded as verifiable assets. — market, concept, knowledge-economy
  • OKF Super Corpus ✅ web — A large-scale example OKF bundle curated by Google Cloud — demonstrates the format at scale across multiple domains. — example, large-scale, google

Guide

  • OKF Blog Post (Chinese) web — Chinese translation of the OKF launch blog post. — blog, translation, chinese

Spec

  • OKF Specification (Chinese) web — Full Chinese translation of the OKF specification, with mandatory requirements and gaps annotated. — spec, translation, chinese
  • OKF Specification (English) ✅ web — Official OKF specification by Google Cloud — v0.2 with sources, trust, lifecycle, and attested computation. — spec, english, official
  • OKF Specification (Official) ✅ web — Google’s official OKF v0.2 specification — the canonical reference for the format. — spec, official, google

Data Model

- id: obsidian-to-okf
  name: obsidian-to-okf
  kind: plugin
  category: document
  url: https://github.com/yzfly/awesome-okf/tree/main/plugins/obsidian-to-okf
  description: Convert Obsidian vaults to OKF — wikilinks become OKF links.
  platform: [python]
  official: false
  tags: [obsidian, wikilink, markdown]

Contributing

Add or edit entries in data/catalog.yaml, then:

awesome-okf validate
awesome-okf readme
pytest

See CONTRIBUTING.md for the entry schema.


🤖 For AI Agents

If you are an AI agent (Claude Code, Codex, Cursor) working on this repo, here’s what you need to know:

Project structure

data/catalog.yaml          # SINGLE SOURCE OF TRUTH — all entries live here
README.template.md         # English template (hand-written narrative)
README.template.zh.md      # Chinese template
README.template.ja.md      # Japanese template
README.template.ko.md      # Korean template
src/awesome_okf/           # CLI + MCP meta-server (Python)
scripts/convert-to-okf.py  # Format converter tool

Rules (non-negotiable)

  1. Never edit README*.md directly — they are generated. Edit data/catalog.yaml, then regenerate.
  2. Never edit the CATALOG blocks in templates — the “ markers are auto-generated.
  3. Adding an entry = append a YAML block to data/catalog.yaml + regenerate + run tests. Schema:
    - id: kebab-case-id          # unique, never reused
      name: Human Readable Name
      kind: tool|plugin|skill|proposal|doc
      category: free-form-group
      url: https://...
      description: One sentence, no marketing fluff.
      platform: [cli|python|web|claude-code]
      official: false            # true ONLY for Google/vendor official
      tags: [3-5 short tags]
  4. Validate before commit: awesome-okf validate && pytest
  5. Never create duplicate ids — edit the existing entry in place.
  6. Multi-language: if you touch a template, mirror the change in all four templates.

Useful commands

awesome-okf stats                    # entry counts by kind
awesome-okf list --kind plugin       # filter by kind
awesome-okf search            # full-text search
awesome-okf validate                 # schema + duplicate-id check
awesome-okf readme                   # regenerate all READMEs
awesome-okf-server                   # MCP meta-server (stdio)

MCP meta-server

The catalog is exposed to agents via awesome-okf-server with four tools: search_catalog, list_catalog, get_catalog_entry, catalog_stats, convert_to_okf. Connect it to your MCP client to query OKF resources programmatically.


  • yzfly/awesome-okf — the original Chinese OKF resource hub
  • OKF Specification — official Google Cloud spec

MIT — see LICENSE. Catalog descriptions link to upstream projects under their respective licenses.

Official distribution

获取与安装

暂未发现可确认的官方软件包地址

当前 README 快照没有出现 npm、PyPI、Crates.io、pub.dev 等官方包页链接。本站不会根据仓库名称猜测下载地址。

本站不托管项目文件;需要安装时,请以项目维护者发布的官方文档为准。

使用前核验

本站保存公开资料用于阅读,不代表安全审计或功能背书。安装前请核对许可证、依赖来源和发布签名,不要直接运行来源不明的二进制文件或高权限脚本。