writing-skills

Create and test agent skills using test-driven development methodology.

92|12|Updated Dec 21, 2025
One-click install
npx skills add https://github.com/ARounder-183/HiFiShifter --skill writing-skills-arounder-183
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: writing-skills
Source: https://github.com/ARounder-183/HiFiShifter/tree/main/.github/skills/writing-skills
Command: npx skills add https://github.com/ARounder-183/HiFiShifter --skill writing-skills-arounder-183

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires graphviz, and includes scripts (resource) and references (resource) components.

What problem does it solve? Writing effective agent skills is hard because untested documentation often fails under real-world pressure. This Skill applies Test-Driven Development (TDD) to skill authoring, ensuring new or edited skills actually change agent behavior before deployment. ## Core Features & Use Cases - TDD-Based Skill Authoring: Follows a RED-GREEN-REFACTOR cycle where you run baseline pressure scenarios with subagents, watch them fail, write the skill, and verify compliance. - Rationalization Bulletproofing: Provides techniques to close loopholes, build rationalization tables, and create red-flag lists so agents cannot excuse rule violations. - Discovery Optimization (CSO): Guidance on writing descriptions, keywords, and naming so future agents can find the skill when it matters. - Use Case: You want to create a new discipline-enforcing skill (e.g., mandatory code review). Use this Skill to design pressure scenarios, test baseline agent behavior, write the skill, and iterate until agents comply under maximum pressure. ## Quick Start Use the writing-skills skill to create a new skill for my project following the TDD methodology.

Frequently Asked Questions about writing-skills

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

FAQPage Schema
How do I create a new agent skill with SKILL.md?

Create a directory containing a SKILL.md file with YAML frontmatter defining name and description, followed by markdown instructions. Follow the TDD cycle: run baseline scenarios without the skill, write minimal content addressing observed failures, then refactor to close loopholes.

How do I test that a skill actually works before deploying it?

Run pressure scenarios with subagents that combine time pressure, sunk cost, and authority. First run without the skill to document baseline failures verbatim, then run with the skill to verify agents comply under maximum pressure.

What should go in the SKILL.md description field?

Write a third-person description starting with 'Use when...' that lists specific triggering conditions and symptoms. Never summarize the skill's workflow in the description, because agents may follow the summary instead of reading the full skill body.

When should I not create a skill?

Skip skill creation for one-off solutions, project-specific conventions better placed in CLAUDE.md, standard practices documented elsewhere, and mechanical constraints that can be enforced with automated validation instead of documentation.

Why do agents ignore my skill even after reading it?

Agents rationalize around rules under pressure using excuses like 'spirit not letter' or 'this case is different.' Counter this by adding explicit negations, rationalization tables, red-flag lists, and foundational principles, then re-test until bulletproof.