skill-creator

Create and iterate GLM-based skills with SKILL.md and trigger evaluations.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/josinesprados-hub/vita-zen --skill skill-creator-josinesprados-hub
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-creator
Source: https://github.com/josinesprados-hub/vita-zen/tree/main/vita-zen-src/vita-zen-main/skills/skill-creator
Command: npx skills add https://github.com/josinesprados-hub/vita-zen --skill skill-creator-josinesprados-hub

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Skill Creator provides a structured workflow and tooling to design, test, and iteratively improve GLM-based skills. It guides users from initial skill drafting through trigger-eval cycles, result interpretation, and benchmarking or description optimization to maximize triggering accuracy and operational effectiveness.

Core Features & Use Cases

  • Define a new skill with a clear objective and executable body in SKILL.md.
  • Run trigger evaluations, review results with an integrated viewer, and identify improvement opportunities.
  • Iterate by rewriting the skill description and tests, then re-run evaluations to converge on a robust, well-triggered skill.
  • Optionally benchmark descriptions and package skills for distribution to teams or users.

Quick Start

Draft a skill's objective, write SKILL.md, and start the trigger-eval-iterate loop to test, review, 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 iterate on a GLM skill from scratch?

To create a GLM skill, you draft a SKILL.md file defining the objective and executable body, then run a trigger-eval-iterate loop to test, review results, and refine the description for accurate triggering.

What is a trigger evaluation in skill development?

A trigger evaluation tests how accurately a skill is invoked by running evaluations against defined tests, allowing you to review results in an integrated viewer and identify opportunities to optimize the skill description.

How do I optimize a skill description to improve triggering accuracy?

You optimize a skill description by rewriting it alongside tests in the SKILL.md file, re-running trigger evaluations, and iterating based on the results to converge on a robust, well-triggered skill.

Can I benchmark and package skills for distribution to a team?

Yes, you can optionally benchmark descriptions across multiple iterations and package the finalized skills to standardize skill development and distribute them to teams or users.

Do I need any specific dependencies to start designing skills?

No specific dependencies are required; you start by drafting the skill's objective and writing the SKILL.md file, which enforces a structured layout with metadata, body, and optional scripts.

What layout structure is enforced when defining a new skill?

Skill creation enforces a structured layout in the SKILL.md file containing metadata, a body, and optional scripts, references, or assets, along with frontmatter validation to standardize development.