Skip to content

natuleadan/skills

Repository files navigation

natuleadan

Natuleadan Skills

Agent skills for AI coding agents across multiple domains

4 count 3 domains MIT License
Skills

1. What is Natuleadan Skills?

A collection of agent skills for AI coding agents spanning multiple domains. Skills are packaged instructions and scripts that extend agent capabilities. Currently focused on programming (Node.js), with biology and cooking domains planned.

Skills follow the Agent Skills format and are installable via npx skills add.

2. Available Skills

Skills use numeric codes to encode a three-level hierarchy: domain, category, and skill. Each skill lives at skills/<skill-code>-<name>/ with a SKILL.md file.

Domain Category Skill Description
01-programming 0101-node 010101-package-security Hardens npm, pnpm, and bun against supply chain attacks
010102-install-and-setup Installs and configures the JS/TS toolchain from scratch
010103-package-ops Day-to-day package management: install, update, audit, publish
0102-agents 010201-skill-creator Guides creation of new Agent Skills

3. Technology Stack

  • skills.sh — Agent skills registry and distribution
  • Python 3 — Skill validation and automation scripts (tools/validate-all.py)
  • Node.js / npx — Skill installation via npx skills add
  • Husky 9 — Git hooks (pre-commit + commit-msg)
  • Commitlint 20 — Conventional commit enforcement
  • Bun — Package manager and runtime (local development only)

4. Project Status

Active domain: 01-programming — 4 production-ready skills for Node.js development: security auditing, toolchain setup, package operations, and agent skill creation.

Placeholder domains: 02-biology and 03-cooking — ready for future skills. Add a new domain by creating skills/NN-name/ with a .gitkeep and registering a new plugin in marketplace.json.

CI/CD: Every push and pull request is automatically validated via GitHub Actions (tools/validate-all.py) to ensure all skills remain compliant.

5. Getting Started

Prerequisites

  • Node.js 18+ — Required for npx skills add (end users)
  • Python 3.8+ — Required for local validation (contributors)
  • Bun — Recommended for local development (optional)

Installation

# Install all skills
npx skills add natuleadan/skills

# Install a specific skill
npx skills add natuleadan/skills --skill 010101-package-security

# Update a skill after changes
npx skills update natuleadan/skills --skill 010101-package-security

Local Setup

git clone https://github.com/natuleadan/skills.git
cd skills
bun install         # Install husky + commitlint
python3 tools/validate-all.py  # Verify everything works

6. Architecture

Skills are structured as a flat registry with a plugin manifest and domain-organized directories. The system has three layers:

  1. Registry.claude-plugin/marketplace.json declares all available skills and their paths. This is the single source of truth for what the repo contains.
  2. Skills — Each skill is a directory under skills/<domain>/ containing SKILL.md (instructions), scripts/ (Python automation), and optionally references/ (supporting documentation).
  3. Validationtools/validate-all.py reads the marketplace manifest, resolves each skill path, and validates every SKILL.md against the Agent Skills format specification.

7. Coding Standards

Commit Messages

Format: type(scope): description

Type Usage
feat New feature or skill
fix Bug fix
upgrade Breaking change
docs Documentation
chore Configuration, tooling

Rules: scope is required, max 100 characters, lowercase subject.

Skill Naming

  • Max 64 characters
  • Lowercase with hyphens only
  • No leading, trailing, or consecutive hyphens
  • Must match directory name (NFKC-normalized)
  • All skills written in English for cross-agent compatibility

Skill Frontmatter (SKILL.md)

Required fields: name, description. Allowed fields: name, description, license, compatibility, metadata, allowed-tools.

8. Testing & Validation

python3 tools/validate-all.py

The validation script performs the following checks:

  • Registry integrity — Every path in marketplace.json resolves to an existing skill directory
  • Frontmatter validation — Custom YAML parser checks required fields and naming rules
  • Directory matching — Skill name must match the containing directory name
  • metadata.json — Every skill must have a valid metadata.json with version, abstract, and references
  • Empty files — Warns on empty .md files in references/

9. Development Workflow

  1. Plan — Define the skill domain, category, and functionality
  2. Create — Scaffold skills/<domain>/<category>/<skill-code>/ with SKILL.md, scripts/, and references/
  3. Register — Add the skill path to .claude-plugin/marketplace.json
  4. Validate — Run python3 tools/validate-all.py to check format compliance
  5. Commit — Stage changes and commit with a conventional message (feat(skills): add new-skill-name). Husky automatically runs validation and commitlint
  6. Push — Push to GitHub; users update via npx skills update natuleadan/skills

10. Contributing

Our Contribution Guidelines document provides detailed conventions for skill creation, covering directory structure, SKILL.md frontmatter rules, naming, validation with python3 tools/validate-all.py, and commit format. All contributions must pass validation before merging. Contribution Guidelines

11. Licensing

MIT — Leonardo Jara

12. Ownership & Copyright

Maintained by Leonardo Jara (@leojara95) under the natuleadan organization (@natuleadan).


Contributors

natuleadan leojara95


Star History

Star History Chart

About

Multi-domain agent skills collection for AI coding agents (Claude, Cursor, Copilot, OpenCode, and more). Covers programming, biology, cooking, and future domains. Installable via npx skills add.

Topics

Resources

License

Contributing

Stars

Watchers

Forks

Contributors