skill-generator

Name: skill-generator
Author: joelhooks

AI Community

Repo stars 679
Author updated Live
Author repo swarm-tools

Fluxly profile Facts only: domain, agents, trust score, runtime, permissions and network

Domain: AI
Compatible agents: Claude Code
Cursor
Cline
Codex
Windsurf
Gemini CLI

+20
Trust score: 88 / 100 · community maintained
Author / version / license: @joelhooks · no license declared
Token usage: Lean
Setup complexity: Guided setup
External API key: Not required
Operating systems: macOS · Linux · Windows
Runtime requirements: Bun · Python
Permissions: Read-only
Write / modify
Shell exec
Network behavior: Local-only
Install commands: 26 variants

Profile is derived at build time from SKILL.md and install vectors. Subject to drift from author intent.

Heads up: 未限定 allowed-tools，默认拥有全部工具权限。

Output preview skill-generator-joelhooks.preview

---
name: skill-generator
description: Meta-skill for generating new skills with proper format and structure. Use when creating new ski…
category: ai
runtime: Bun / Python
---

# skill-generator output preview

## PART A: Task fit
- Use case: Meta-skill for generating new skills with proper format and structure. Use when creating new skills for the swarm system or when agents need to generate skill scaffolds. Ensures skills follow conventions (frontmatter format, directory structure, bundled resources)..
- Inputs: target material, constraints, expected output, and acceptance criteria.
- Evidence boundary: follow “Quick Start / Skill Format Conventions / Frontmatter Requirements” and do not present inference as author intent.

## PART B: Execution result
- **01** The card summarizes the use case; runtime output centers on “Meta-skill for generating new skills with proper format and structure. Use when creating new skills for the swarm system or when agents need to generate skill scaffolds. Ensures skills follow conventions (frontmatter format, directory structure, bundled resources).”.
- **02** When the source has headings, the agent prioritizes “Quick Start / Skill Format Conventions / Frontmatter Requirements” so the result follows the author’s structure.
- **03** Typical output includes task judgment, concrete steps, required commands or file edits, validation, and follow-up options.
- **04** Risk context follows the fingerprint: read files, write/modify files, run shell commands; mostly runs locally; usually needs no extra API key.

## Running Rules
- read files, write/modify files, run shell commands; mostly runs locally; usually needs no extra API key.
- Validate with a small sample before expanding scope.
- Return the result, validation criteria, and next iteration options.

SKILL.md SKILL.md

---
name: skill-generator
description: Meta-skill for generating new skills with proper format and structure. Use when creating new ski…
category: ai
source: joelhooks/swarm-tools
---

# skill-generator

## When to use
- Meta-skill for generating new skills with proper format and structure. Use when creating new skills for the swarm syst…
- Use it when the task has clear inputs, repeatable steps, and validation criteria.

## What to provide
- Target material, scope, expected result, and forbidden changes.
- Whether network, commands, file writes, or external services are allowed.

## Execution rules
- Organize steps around “Quick Start / Skill Format Conventions / Frontmatter Requirements” and keep inference separate from source facts.
- read files, write/modify files, run shell commands; mostly runs locally; usually needs no extra API key.
- Validate with a small sample before expanding the task.

## Output requirements
- Return the deliverable, key evidence, validation method, and next action.
- Mark missing information as unknown; do not invent commands, platforms, or dependencies.

Evidence boundary and execution chain

The author source anchors workflow facts; repository files anchor sources and commands; Fluxly only adds fit, limitations, and quality judgment.

skill "skill-generator" {
  input      -> user goal + target files + boundaries + acceptance criteria
  context    -> Quick Start / Skill Format Conventions / Frontmatter Requirements
  rules      -> SKILL.md triggers / order / output contract
  runtime    -> Bun / Python | read files, write/modify files, run shell commands | mostly runs locally
  guardrails -> usually needs no extra API key + small-sample validation + diff/log review
  output     -> copyable result + checklist + next iteration
}

Interpretation is structured for decision-making; original keeps the upstream SKILL.md unchanged.

Decide Fit First

Core job: Meta-skill for generating new skills with proper format and structure. Use when creating new skills for the swarm system or when…
Best fit: Use it when the task has reusable inputs, steps, and validation criteria rather than a one-off answer.
Avoid forcing it: If the source lacks commands, platform support, or external-service evidence, keep those fields unknown instead of guessing.

Design Intent

Structure: The skill is organized around “Quick Start”, “Skill Format Conventions”, “Frontmatter Requirements”, “Directory Structure”, showing how the author expects the agent to judge fit, collect context, and produce verifiable output.
Trigger evidence: Prioritize the author’s wording around when to use it, what context to collect, and what output shape to produce.
Evidence boundary: Author text states facts, repository files prove commands and paths, and Fluxly only adds fit, limits, and usage judgment.

How To Use It

Inputs: Provide target material, scope, expected result, forbidden changes, and validation method.
Invocation: Name skill-generator directly; if the source includes slash commands, start with the command and then add task context.
Validation: Start small and check whether the result follows “Quick Start / Skill Format Conventions / Frontmatter Requirements” before expanding.

Boundaries And Review

Dependencies: It usually needs no extra API key, so start with a small validation task.
Permissions: Declared permissions include read / write / shell-exec; ask the agent to state file, command, and rollback boundaries before acting.
Quality bar: A useful result names the deliverable, evidence, and next action. Generic prose means the task needs tighter context.

Skill Generator

Generate new skills with proper format, structure, and conventions. This meta-skill helps agents create skills without hallucinating the format.

Quick Start

To generate a new skill:

bash scripts/generate-skill.sh <skill-name> [target-directory]

This creates a complete skill scaffold with:

SKILL.md with proper frontmatter
scripts/ directory for executable helpers
references/ directory for documentation
Placeholder content following conventions

Skill Format Conventions

Every skill MUST include:

SKILL.md (required) - Main skill file with:
- YAML frontmatter (name, description)
- Markdown body with instructions
Bundled Resources (optional):
- scripts/ - Executable code (bash/python/etc)
- references/ - Documentation loaded on-demand
- assets/ - Files used in output (templates, etc)

Frontmatter Requirements

---
name: skill-name
description: What the skill does AND when to use it. Include triggering scenarios.
---

The description field is critical for skill discovery and triggering. Include:

What the skill does
When to use it (specific triggers)
What contexts activate it

Directory Structure

skill-name/
├── SKILL.md (required)
├── scripts/ (optional)
│   └── example-script.sh
├── references/ (optional)
│   └── conventions.md
└── assets/ (optional)
    └── template-file

Writing Effective Skills

Keep SKILL.md Lean

Target <500 lines in SKILL.md. Move detailed content to references/:

Core workflow → SKILL.md
Detailed examples → references/
API docs → references/
Long explanations → references/

Use Imperative Form

Write instructions as commands:

"Read the file first" ✓
"You should read the file" ✗
"Check for patterns" ✓
"Consider checking patterns" ✗

Progressive Disclosure

Skills use three-level loading:

Metadata (~100 words) - Always in context
SKILL.md body (<5k words) - When skill triggers
Bundled resources (unlimited) - Loaded as needed

Bundled Resources

scripts/

Executable code for deterministic tasks:

When the same code is rewritten repeatedly
When reliability is critical
Run via bash/python without loading to context

Make scripts executable:

chmod +x scripts/my-script.sh

references/

Documentation loaded on-demand:

Database schemas
API documentation
Detailed workflow guides
Domain knowledge

Keep reference files focused. For files >100 lines, include a table of contents.

Reference from SKILL.md with clear guidance on when to read:

See references/api-docs.md for complete API reference.

assets/

Files used in output (not loaded to context):

Templates
Images/icons
Boilerplate code
Fonts/typography

What NOT to Include

Do NOT create these files:

README.md
INSTALLATION_GUIDE.md
QUICK_REFERENCE.md
CHANGELOG.md

Skills should contain only what an AI agent needs to execute the task. No auxiliary documentation.

Validation

Before finalizing, validate the skill:

bun scripts/validate-skill.ts path/to/skill

Checks:

YAML frontmatter format
Required fields present
No TODO placeholders
No extraneous files
Naming conventions

Reference

See references/conventions.md for complete skill format specification.

Install command

One-click installPick a runner, click Copy, paste into your terminal, press enter

npx skills add joelhooks/swarm-tools/packages/opencode-swarm-plugin/global-skills/skill-generator -y

Open general install guide

View skill source

Works with

Claude Code
Cursor
Cline
Codex
Windsurf
Gemini
OpenCode
Roo Code
AAide
+Augment
GitHub Copilot
OpenClaw
KKiro
Junie
WWarp
Goose
Trae
Kilo Code
OpenHands
Amp
Zencoder
CContinue
AIAider
CodeBuddy
CodeGeeX
Replit Agent

Security audit

88 / 100

Community

未声明 license/许可。
未声明 version。
未限定 allowed-tools，默认拥有全部工具权限。

Metadata

Source: github
joelhooks/swarm-tools
中文

More by @joelhooks

S recall 98

Category & tags

Category AI