By catancs
Teach coding agents to use the Open Knowledge Format for portable, structured knowledge
The Open Knowledge Format skill for coding agents.
Teaches AI agents to store, retrieve, and manage knowledge using OKF, an open specification from Google.
Foundation models are powerful, but they lack context. The knowledge agents need (table schemas, API docs, runbooks, metrics) lives in fragmented, incompatible systems. Every agent builder solves the same context-assembly problem from scratch.
OKF is Google's answer: a minimal, vendor-neutral format where knowledge lives as markdown files with YAML frontmatter, organized in directories, cross-linked, version-controllable, and readable by both humans and agents.
This skill makes your coding agent use OKF by default, so knowledge is structured, portable, and compounds across sessions.
flowchart LR
subgraph WITHOUT[" Without OKF"]
direction TB
A1[Agent documents database] --> A2[random markdown, no conventions]
A2 --> A3[next agent starts from scratch]
A3 --> A4[knowledge lost across sessions]
end
subgraph WITH[" With OKF"]
direction TB
B1[Agent documents database] --> B2[OKF bundle with structured frontmatter]
B2 --> B3[next agent reads the bundle]
B3 --> B4[knowledge versioned, reviewable, portable]
end
style WITHOUT fill:#1a1a2e,stroke:#e74c3c,color:#fff
style WITH fill:#1a1a2e,stroke:#2ecc71,color:#fff
You say:
"Document my Postgres orders table as an OKF bundle"
Your agent produces:
knowledge/
├── index.md
├── datasets/
│ ├── index.md
│ └── sales.md
└── tables/
├── index.md
├── orders.md <-- this file
└── customers.md
---
type: Table
title: Orders
description: One row per completed customer order
tags: [sales, orders]
timestamp: 2026-06-17T00:00:00Z
---
# Schema
| Column | Type | Description |
|--------------|--------|--------------------------------------|
| order_id | STRING | Unique order identifier |
| customer_id | STRING | FK to customers |
| total_usd | NUMERIC| Order total in USD |
| placed_at | TIMESTAMP | When the order was placed |
# Related
See customers.md for the join key.
What just happened:
flowchart LR
S1["1. Create directory"] --> S2["2. Generate YAML frontmatter"]
S2 --> S3["3. Write schema table"]
S3 --> S4["4. Cross-link concepts"]
S4 --> S5["5. Ready for any agent"]
style S1 fill:#2d3436,stroke:#0984e3,color:#fff
style S2 fill:#2d3436,stroke:#0984e3,color:#fff
style S3 fill:#2d3436,stroke:#0984e3,color:#fff
style S4 fill:#2d3436,stroke:#0984e3,color:#fff
style S5 fill:#2d3436,stroke:#00b894,color:#fff
From within Claude Code:
/plugin marketplace add catancs/okf-skill
/plugin install okf-skill@okf-skill
Open source. Use it, fork it, improve it.
If the OKF spec evolves, update the skill. If you find patterns that work better, share them. The value of a knowledge format comes from adoption, not ownership.
Built by @catancs . MIT licensed
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
npx claudepluginhub catancs/okf-skill --plugin okf-skillClaude audits its own context window, finds which MCP servers & skills tax every turn or sit unused, and disables them safely
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Design fluency for frontend development. 1 skill with 23 commands (/impeccable polish, /impeccable audit, /impeccable critique, etc.) and curated anti-pattern detection.
Behavioral guidelines to reduce common LLM coding mistakes, derived from Andrej Karpathy's observations on LLM coding pitfalls
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.