skill-creator

Create and iteratively refine AI skills with trigger logic and evaluation workflows.

7|2|Updated Apr 30, 2026
One-click install
npx skills add https://github.com/MRWillisT/PullNexus --skill skill-creator-mrwillist
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/MRWillisT/PullNexus/tree/main/skills/skill-creator
Command: npx skills add https://github.com/MRWillisT/PullNexus --skill skill-creator-mrwillist

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The skill-creator helps you turn an idea or an existing draft into a reusable, high-quality skill by guiding iterative drafting, test creation, evaluation, and description tuning.

Core Features & Use Cases

  • Skill authoring loop: Draft a skill from your intent, clarify trigger conditions, and refine instructions based on results.
  • Evaluation-driven improvement: Generate a test set, run comparisons (with-skill vs baseline), draft measurable assertions, and review outcomes to reduce underperformance.
  • Description optimization: Improve triggering accuracy so the skill activates reliably when users need it, not just when it matches obvious keywords.

Quick Start

Use skill-creator when you want to create a new skill or improve an existing one, and ask an AI to help you draft the skill, build a small test set, run evaluations, and iterate until triggering and outputs look correct.

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 iterate on AI skills with evaluation-driven refinement?

To create AI skills with evaluation-driven refinement, draft instructions from user intent, define trigger contexts, generate test prompts with assertions, run benchmark comparisons, and grade outcomes to reduce underperformance.

What is the best way to optimize skill descriptions for better trigger activation accuracy?

Optimizing skill descriptions for trigger accuracy involves tuning the SKILL.md text so the skill activates reliably based on contextual user intent rather than matching only obvious keywords.

How does benchmarking skill performance against a baseline work?

Benchmarking skill performance involves running comparison workflows with the skill versus a baseline, grading measurable assertions on test prompts, and aggregating results to identify and correct output deficiencies.

Can I use this to improve an existing skill draft rather than starting from scratch?

Yes, you can enhance existing skill descriptions by clarifying trigger conditions, refining instructions based on evaluation results, and iterating until triggering logic and outputs meet your quality standards.

Why do my custom AI skills fail to trigger correctly in context?

Custom AI skills fail to trigger correctly when the SKILL.md description lacks precise contextual logic, requiring description optimization to align activation with actual user intent instead of superficial keyword matching.

Do I need predefined test sets to run skill evaluation workflows?

No, you do not need predefined test sets; the skill authoring loop can generate a test set, draft measurable assertions, and run comparative evaluations to iteratively refine the skill.