skills/skill-creator/SKILL.md
Creates new AI agent skills following the Agent Skills spec. Trigger: When user asks to create a new skill, add agent instructions, or document patterns for AI.
npx skillsauth add 333-333-333/agents skill-creatorInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Create a skill when:
Don't create a skill when:
skills/{skill-name}/
├── SKILL.md # Required - main skill file
├── assets/ # Optional - templates, schemas, examples
│ ├── template.py
│ └── schema.json
└── references/ # Optional - links to local docs
└── docs.md # Points to docs/developer-guide/*.mdx
---
name: {skill-name}
description: >
{One-line description of what this skill does}.
Trigger: {When the AI should load this skill}.
metadata:
author: 333-333-333
version: "1.0"
type: {generic|project|meta}
scope: [{root|directory-name}]
auto_invoke:
- "{Action that triggers this skill}"
- "{Another action that triggers this skill}"
---
## When to Use
{Bullet points of when to use this skill}
## Critical Patterns
{The most important rules - what AI MUST know}
## Code Examples
{Minimal, focused examples}
## Commands
```bash
{Common commands}
---
## Naming Conventions
| Type | Pattern | Examples |
|------|---------|----------|
| Generic skill | `{technology}` or `{tech}-{concern}` | `typescript`, `go-tdd`, `go-gin-handlers` |
| Project-specific | `{project}-{component}` | `bastet-booking`, `bastet-auth` |
| Workflow skill | `{action}-{target}` | `skill-creator`, `skill-sync` |
| Meta skill | `skill-{action}` | `skill-creator`, `skill-sync` |
---
## Decision: Skill Type
Patterns apply to ANY project? → type: generic Patterns are specific to THIS repo? → type: project Skill manages the skills system? → type: meta
**Important**: The `type` field determines where the skill appears in AGENTS.md:
- `generic` → **Generic Skills** table (reusable across projects)
- `project` → **Project Skills** table (repo-specific)
- `meta` → **Meta-Skills** section (manually managed, not synced)
---
## Decision: assets/ vs references/
Need code templates? → assets/ Need JSON schemas? → assets/ Need example configs? → assets/ Link to existing docs? → references/ Link to external guides? → references/ (with local path)
**Key Rule**: `references/` should point to LOCAL files (`docs/developer-guide/*.mdx`), not web URLs.
---
## Frontmatter Fields
| Field | Required | Description |
|-------|----------|-------------|
| `name` | Yes | Skill identifier (lowercase, hyphens) |
| `description` | Yes | What + Trigger in one block |
| `metadata.author` | Yes | `333-333-333` |
| `metadata.version` | Yes | Semantic version as string |
| `metadata.type` | Yes | `generic`, `project`, or `meta` |
| `metadata.scope` | Yes | Array of scopes: `[root]`, `[api]`, `[root, api]`, etc. |
| `metadata.auto_invoke` | Yes | String or list of actions that trigger this skill |
### Scope Values
| Scope | Updates |
|-------|---------|
| `root` | `./AGENTS.md` (repository root) |
| `{directory}` | `./{directory}/AGENTS.md` |
A skill can target multiple scopes: `scope: [root, api, web]`
### Auto-invoke Format
Single action:
```yaml
auto_invoke: "Creating new components"
Multiple actions:
auto_invoke:
- "Creating new components"
- "Refactoring component structure"
After creating the skill, run the sync script to register it in AGENTS.md:
./skills/skill-sync/assets/sync.sh
This automatically updates the skill tables and auto-invoke sections based on metadata.type, metadata.scope, and metadata.auto_invoke.
skills/)name and description (with Trigger)metadata.type (generic/project/meta)metadata.scope (where to register)metadata.auto_invoke (when to trigger)./skills/skill-sync/assets/sync.shtesting
Review Flutter components and screens for UX/UI compliance. Trigger: When user invokes /ux-review command or requests UX audit.
development
TypeScript strict patterns and best practices. Trigger: When implementing or refactoring TypeScript in .ts/.tsx (types, interfaces, generics, const maps, type guards, removing any, tightening unknown).
testing
Testing philosophy and strategy for every feature: test pyramid, mandatory levels per change type, completion checklist, and skill delegation. Trigger: When planning tests for a feature, reviewing test coverage, defining acceptance criteria, or asking what tests a change needs.
development
Terraform security practices: sensitive variables, secret management, state protection, .gitignore patterns, and CI/CD credential handling. Trigger: When handling secrets in Terraform, configuring state backends, reviewing .gitignore for Terraform, or setting up CI/CD pipelines for infrastructure.