skills.rest
joshsymonds
joshsymondsCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

writing-skills

Create and modify Claude skills with evaluation-driven testing cycles.

Updated Jan 23, 2026
Search Tags:#skill development#testing#tdd#refactoring#claude skills#evaluation
One-click install
npx skills add https://github.com/joshsymonds/gambit --skill writing-skills-joshsymonds
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: writing-skills
Source: https://github.com/joshsymonds/gambit/tree/main/skills/writing-skills
Command: npx skills add https://github.com/joshsymonds/gambit --skill writing-skills-joshsymonds

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill streamlines the creation and modification of Claude skills by enforcing a rigorous evaluation-driven development process, ensuring skills are robust, effective, and meet specific criteria before finalization.

Core Features & Use Cases

  • Evaluation-Driven Development: Guides users through a TDD-like cycle (RED-GREEN-REFACTOR) using subagents for testing.
  • Rigorous Testing: Includes baseline tests, pressure tests (time, sunk cost, authority), multi-model tests, and triggering tests.
  • Use Case: When you need to create a new skill for managing customer support tickets, or modify an existing skill to handle new types of inquiries, use this skill to ensure it's well-defined, tested, and triggers correctly.

Quick Start

Use the writing-skills skill to create a new skill for summarizing meeting notes.

Frequently Asked Questions about writing-skills

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

FAQPage Schema
How do I test Claude skills using an evaluation-driven development methodology?▼

You can create and modify Claude skills by defining requirements, then iterating through RED-GREEN-REFACTOR testing cycles with subagents until the skill passes all evaluation criteria.

What types of tests should I run when developing Claude skills?▼

Yes, TDD principles apply to Claude skill development by using subagents to execute RED-GREEN-REFACTOR cycles, enforcing strict testing protocols before a skill is finalized.

How do I ensure a Claude skill triggers correctly for specific use cases?▼

Triggering tests validate that a Claude skill activates appropriately for its intended use cases, verifying the skill definition and criteria prevent misfires before deployment.

What is the best way to refine an existing Claude skill to handle new inquiry types?▼

Refining an existing Claude skill requires defining the new inquiry requirements, then running iterative evaluation-driven tests with subagents to validate the updated skill behavior and triggering logic.

Do I need subagents to evaluate Claude skill quality?▼

Subagents are used within the evaluation-driven development process to independently execute baseline, pressure, and multi-model tests, providing objective validation of Claude skill quality and reliability.

Related Skills

Explore More Software Engineering Skills→
ysyecustysyecust
40

eval-harness

Codify and automate Claude Code session evaluations with pass@k metrics.

Community
Intermediate
accolveraccolver
4

skill-maker

Create and refine agent skills through scripted eval loops.

Community
Advanced
seanGSISGseanGSISG
3

testing-skills-with-subagents

Test skills with subagents to verify resistance to rationalization under pressure.

Community
Basic
GameDevFoxGameDevFox

skill-creator

Draft SKILL.md content and run with-skill versus without-skill trigger tests.

Community
Advanced