skill-creator

Create, validate and iterate AI skills with structured SKILL.md drafts and YAML frontmatter.

512|64|Updated Mar 13, 2025
One-click install
npx skills add https://github.com/agnt-gg/agnt --skill skill-creator-agnt-gg
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/agnt-gg/agnt/tree/main/backend/skills/skill-creator
Command: npx skills add https://github.com/agnt-gg/agnt --skill skill-creator-agnt-gg

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables teams to design, test, and iteratively improve AI skills, reducing time-to-value and ensuring consistent quality across skill development.

Core Features & Use Cases

  • Draft new skills by capturing intent, trigger contexts, and success criteria.
  • Generate, run, and analyze test prompts to evaluate performance.
  • Iterate rapidly with automated feedback loops, eval viewers, and suggested refinements to trigger phrases and descriptions.

Quick Start

Draft a new skill by telling me its goal, how you want it triggered, and any constraints, and I’ll produce a complete SKILL.md and a plan to test and refine it.

Frequently Asked Questions about skill-creator

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

FAQPage Schema
How do I create and test AI skills iteratively?

To create and test AI skills, define the goal, trigger contexts, and success criteria to generate a SKILL.md file, then run test prompts and gather feedback for iterative improvements.

What is a SKILL.md file and how does it work for automation workflows?

A SKILL.md file defines an AI skill's purpose, trigger conditions, and evaluation criteria, requiring YAML frontmatter with name and description to enable structured automation workflows and safety reviews.

How do I write trigger conditions and evaluation criteria for a new skill?

Write trigger conditions and evaluation criteria by capturing the specific intent and constraints of your task, which establishes when the skill activates and how its performance is measured during testing.

Does skill creation require a specific format or safety review process?

Skill creation requires a YAML frontmatter block containing the name and description, supports optional resources directories, and mandates a safety review process to prevent potential misuse of the automation.

What's the best way to validate AI skill performance before deployment?

The best way to validate AI skill performance is by generating and running proposed test prompts, analyzing the results through an eval viewer, and applying suggested refinements to trigger phrases and descriptions.