skills.rest
Probably-Group
Probably-GroupOFFICIAL
Explore all skills by this author →
Website𝕏@problynotingGitHub

skill-creation-expert

Generate Claude Code SKILL.md files with YAML frontmatter and CWE-based security rules.

3|1|Updated Dec 3, 2025
Search Tags:#claude code#skill design#anti-hallucination#prompt templating#cwe security#code generation rules#skill_loader
One-click install
npx skills add https://github.com/Probably-Group/Dev-AID --skill skill-creation-expert
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: skill-creation-expert
Source: https://github.com/Probably-Group/Dev-AID/tree/main/.dev-aid/skills/expert/skill-creation-expert
Command: npx skills add https://github.com/Probably-Group/Dev-AID --skill skill-creation-expert

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It solves the challenge of creating reliable, anti-hallucination Claude Code skills that follow a consistent SKILL.md format while enforcing secure coding patterns instead of vague advice.

Core Features & Use Cases

  • Skill template enforcement: Provides the required SKILL.md structure so prompts become actionable operational instructions.
  • Anti-hallucination security rules: For technical skills, mandates technology-specific CWE-style Do not/Instead patterns with short correct code examples.
  • Quality trimming strategy: Guides what to include vs. omit (e.g., move large code/tests into references) to keep skills focused and usable.

Quick Start

Use this skill to generate and validate a new SKILL.md for your next Claude Code skill draft, ensuring it includes name/description frontmatter and the correct security-pattern section when technical.

Frequently Asked Questions about skill-creation-expert

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

FAQPage Schema
How do I create a SKILL.md file for Claude Code that prevents AI hallucination?▼

To create an anti-hallucination SKILL.md for Claude Code, use a structured template enforcing actionable operational instructions and technology-specific CWE-style Do not/Instead security rules with short correct code examples.

What is the correct YAML frontmatter format for a Claude Code skill?▼

The correct YAML frontmatter for a Claude Code skill requires including the name and description fields to satisfy discovery and activation requirements before defining the actionable, trimmed instruction body.

How do I write CWE-based security rules for code generation prompts?▼

Write CWE-based security rules by defining technology-specific Do not/Instead patterns that contrast vulnerable code with short correct examples, ensuring secure coding patterns are enforced during code generation.

What should I exclude from a SKILL.md template to keep Claude Code prompts focused?▼

To keep SKILL.md templates focused, apply a quality trimming strategy by moving large code blocks and tests into separate references, omitting them from the main instruction body to maintain usability.

Does my Claude Code skill need security guardrails if it is not generating code?▼

Security guardrails mandating CWE-style Do not/Instead patterns are specifically required for technical skills, meaning non-technical skills can omit this section while still following the standard SKILL.md format.

Why does my Claude Code skill produce unreliable AI behavior and inconsistent outputs?▼

Unreliable AI behavior and inconsistent outputs occur when skills lack a consistent SKILL.md format, actionable operational instructions, and anti-hallucination constraints like technology-specific security patterns.

Related Skills

Explore More Software Engineering Skills→
sunLeeesunLeee

quality-bandit

Configure Bandit security rules and CWE/OWASP mappings for Python projects.

Community
Intermediate
Svenja-devSvenja-dev
38

prompt-architect

Converts vague requirements into structured, best-practice prompts for Claude 4.x.

Community
Advanced
doruciocleadorucioclea
1

vibeship-security-writer

Generate SEO-optimized cybersecurity articles about AI-generated code vulnerabilities.

Community
Advanced
devCharuzudevCharuzu

senior-prompt-engineer

Design prompts and evaluation frameworks for LLM workflows.

Community
Advanced