skill-creator

Guide creation and validation of modular Codex skills with SKILL.md frontmatter.

1|Updated Apr 4, 2026
One-click install
npx skills add https://github.com/p-potvin/vaultwares-pipelines --skill skill-creator-p-potvin
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/p-potvin/vaultwares-pipelines/tree/main/.github/agents/codex/.system/skill-creator
Command: npx skills add https://github.com/p-potvin/vaultwares-pipelines --skill skill-creator-p-potvin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This guide removes uncertainty and repetition from building Codex Skill Units by describing a minimal, discoverable SKILL.md, resource organization, and validation steps so authors can produce consistent, efficient skills that trigger reliably.

Core Features & Use Cases

  • Clear triggering metadata: Specifies the required SKILL.md frontmatter (name and description) and how to phrase triggers.
  • Progressive disclosure & resource patterns: Guidance for keeping SKILL.md concise while using scripts, references, and assets on-demand.
  • Practical workflow: Step-by-step instructions for initializing, editing, validating, and forward-testing skills to ensure reliable activation and low context overhead.
  • Use Case: A developer needs a repeatable pattern to add a PDF-processing skill with deterministic scripts and searchable references without bloating the model context.

Quick Start

Initialize a new skill using the init_skill.py generator, write a concise SKILL.md frontmatter with name and description, and add scripts, references, or assets only when they are truly reusable.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I create production-ready Codex skills with proper metadata and validation?

To create Codex skills, initialize a new skill using the init generator, write a concise SKILL.md frontmatter with a name and description, and add scripts, references, or assets only when truly reusable. Validate metadata and behavior before deployment to ensure reliable activation.

What is progressive disclosure in Codex skills and how does it minimize context bloat?

Progressive disclosure in Codex skills keeps the SKILL.md file concise by loading scripts, references, and assets on-demand rather than upfront. This approach minimizes context bloat by ensuring the model only processes resources when they are truly needed for the task.

How do I structure scripts, references, and assets when building a modular skill?

Structure scripts, references, and assets by organizing them as modular on-demand resources alongside a concise SKILL.md file. Add these components only when they are truly reusable, following naming rules and resource patterns to ensure deterministic execution without bloating model context.

What are the required frontmatter fields for a SKILL.md file to trigger reliably?

The required frontmatter fields for a SKILL.md file are name and description. Specifying these triggering metadata fields ensures the skill is discoverable and activates reliably, providing a minimal and consistent pattern for authors to produce efficient skills.

What is the best way to validate skill metadata and behavior before deployment?

The best way to validate skill metadata and behavior is to follow the specified validation steps and forward-test the skill. This ensures the SKILL.md contains the required fields and that the skill triggers reliably with low context overhead before deployment.

When should I not add scripts or references to my SKILL.md?

You should not add scripts or references to your SKILL.md when they are not truly reusable. Keeping resources out of the core file until needed follows progressive disclosure patterns and minimizes context bloat, ensuring efficient skill activation.