skill-creator_skill

This skill helps you create and update modular Antigravity agent skills, packaging them for distribution and ensuring compliance with skill standards.

Python

60

GitHub Stars

1

Bundled Files

2 months ago

Catalog Refreshed

4 months ago

First Indexed

Readme & install

Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.

Installation

Preview and clipboard use veilstrat where the catalogue uses aiagentskills.

npx veilstrat add skill kienhaminh/anti-chaotic --skill skill-creator

SKILL.md9.0 KB

Overview

This skill helps create, update, and package modular Agent Skills for the Antigravity/Gemini CLI. It provides standards, templates, and automation helpers so agents and developers can produce consistent, version-controlled skill packages ready for distribution.

How this skill works

The skill inspects project layout and creates a canonical skill directory with metadata, scripts, references, and assets. It validates naming, frontmatter metadata, token sizing, and recommended files, then runs quick validation and produces a change report. It also generates domain-specific question artifacts so you can collect the exact requirements before implementation.

When to use it

Creating a new agent skill for a project-specific or global catalog
Upgrading an existing skill to add capabilities, scripts, or documentation
Packaging a skill for distribution or personal/global installation
Validating skill structure, metadata, and compatibility before release
Generating a focused questionnaire when requirements are unclear

Best practices

Always clarify scope first using domain-specific questions before coding or packaging
Use lowercase hyphenated names and keep descriptions concise with explicit triggers and keywords
Keep instructional content compact (<5000 tokens) and separate knowledge from process
Provide executable scripts with error handling and a requirements file for dependencies
Store reusable references and large resources in dedicated directories loaded on demand
Run automated validation and produce a change report after each update

Example use cases

Initialize a new frontend skill that provides linting, component templates, and design tokens
Upgrade an AI-modeling skill to include inference scripts and updated API references
Package a DevOps skill with deployment scripts and asset templates for team-wide distribution
Audit an existing skill for token bloat, missing metadata, or incorrect naming conventions
Create a custom questionnaire to capture constraints for a backend-focused skill

FAQ

I always gather scope, tools/APIs, expected inputs/outputs, and any existing patterns or workflows specific to the domain.

How do you ensure naming and metadata are valid?

The skill enforces lowercase hyphenated names, checks the directory name match, and validates frontmatter fields and token sizing before packaging.

Can you automate validation and reporting?

Yes — after changes I run quick validation and generate a structured change report showing files changed, token counts, and validation results.