writing-skills

Create and refine SKILL.md documentation with YAML frontmatter for agent workflows.

1|Updated Apr 27, 2026
One-click install
npx skills add https://github.com/darkmatter/skills --skill writing-skills-darkmatter
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: writing-skills
Source: https://github.com/darkmatter/skills/tree/main/skills/writing-skills
Command: npx skills add https://github.com/darkmatter/skills --skill writing-skills-darkmatter

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps create, revise, and verify Skills that are discoverable, concise, and resilient under real agent pressure.

Core Features & Use Cases

  • Skill Authoring: Shape SKILL.md frontmatter and body content so other agents can find and use the skill correctly.
  • Pressure Testing: Design baseline scenarios, observe failures, and harden instructions against rationalizations and loopholes.
  • Documentation Craft: Organize supporting references, examples, and structure for reusable process guidance.
  • Use Case: You are turning a messy internal workflow into a reliable Skill that survives deadlines, sunk cost, and other compliance pressures.

Quick Start

Use this skill to draft or improve a Skill, then test it on realistic pressure scenarios before deploying it.

Frequently Asked Questions about writing-skills

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

FAQPage Schema
How do I write agent skills that AI models actually follow during execution?

To write agent skills that models follow, structure your SKILL.md with concise YAML frontmatter and test-first validation to ensure reliable reuse. This approach applies rationalization hardening to prevent agents from bypassing instructions under compliance pressure.

What is TDD for prompt engineering and skill testing?

TDD for prompt engineering applies the red-green-refactor cycle to skill authoring by designing baseline scenarios, observing agent failures, and hardening instructions. This test-first validation ensures your skill survives real agent pressure before deployment.

Can I use the same Skill documentation across Claude Code, Codex, and OpenCode?

Yes, you can deploy the same Skill documentation across Claude Code, Codex, and OpenCode. The skill authoring process requires structured SKILL.md content with concise YAML frontmatter so agents across these tools can discover and follow the workflow.

Why does my AI agent ignore skill instructions when facing deadlines or sunk cost?

Agents ignore skill instructions under compliance pressures like deadlines due to rationalizations and loopholes in the documentation. Pressure testing your skills against realistic scenarios hardens the instructions, preventing rationalization failures during execution.

What's the best way to structure SKILL.md content for agent discoverability?

The best way to structure SKILL.md for agent discoverability is organizing concise YAML frontmatter with clear metadata, followed by a body containing supporting references and examples. This structured content allows agents to reliably find and execute the intended workflow.

How do I pressure test an AI agent skill before deployment?

Pressure test an AI agent skill by designing baseline scenarios, observing agent failures, and hardening instructions against rationalizations and loopholes. This test-first validation catches compliance failures before deploying across agent tools.