writing-skills

Codifies process knowledge into reusable skill documentation with test-first development.

3|2|Updated Oct 10, 2025
One-click install
npx skills add https://github.com/liauw-media/CodeAssist --skill writing-skills-liauw-media
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: writing-skills
Source: https://github.com/liauw-media/CodeAssist/tree/main/skills/meta/writing-skills
Command: npx skills add https://github.com/liauw-media/CodeAssist --skill writing-skills-liauw-media

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides a structured, test-driven approach to creating high-quality, reusable AI process documentation, ensuring that every skill is effective, clear, and addresses real-world problems. This codifies tacit knowledge and promotes consistent practices.

Core Features & Use Cases

  • TDD for Process Documentation: Guides you through a RED (baseline failures) → GREEN (write skill) → REFACTOR (test and refine) cycle to ensure skills solve actual problems.
  • Structured Skill Design: Defines four types of skills (Discipline, Technique, Pattern, Reference) and provides a comprehensive document structure for clarity and completeness.
  • Use Case: When noticing developers repeatedly making the same testing mistakes, this skill will guide you to document those failures (RED), then create a testing-anti-patterns skill with clear rules and examples (GREEN), and finally test and refine it to ensure it prevents those mistakes (REFACTOR).

Quick Start

I've identified a common problem in our workflow. Use the writing-skills skill to help me create a new AI skill to address it.

Frequently Asked Questions about writing-skills

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I document process knowledge so my team can reuse workflows consistently?

Process documentation codifies tacit knowledge into reusable skill formats by capturing repeatable workflows, standardizing them with clear structure (core principle, steps, examples, edge cases), and sharing them across teams. This ensures consistent practices and enables knowledge transfer without repeated explanation.

What's the best way to create AI process documentation that actually solves problems?

Use test-driven documentation: first document baseline failures (RED), then write the skill to address them (GREEN), then test and refine it (REFACTOR). This cycle ensures your documentation solves real problems and prevents repeated mistakes.

How do I structure a skill to make it clear and effective for my team?

Structure skills using four types (Discipline, Technique, Pattern, Reference) with a comprehensive document format including core principle, process steps, concrete examples, and edge cases. This standardized structure ensures clarity, completeness, and cross-model applicability.

Can I use this approach to capture common mistakes and turn them into best practices?

Yes. When you notice repeated errors in your workflow, document those failures as baseline test cases, then create a skill with clear rules and examples to prevent them, and refine it through testing. This converts tacit warnings into actionable documented patterns.

What counts as a reusable skill worth documenting?

A reusable skill addresses a repeatable workflow or common problem your team encounters multiple times. It should have discoverable patterns, clear steps, and applicability across contexts—such as testing anti-patterns, onboarding procedures, or community contribution workflows.

Do I need existing documentation or tools to start creating a skill?

No. Start with observed problems in your workflow. Identify where developers struggle, document those baseline failures, then build the skill around solving them. The test-first approach works with minimal prerequisites and builds documentation iteratively.