Skills

A collection of reusable skills for Claude Code — each skill guides Claude's behaviour through structured markdown and reference templates. No executable code — skills are documentation-driven, teaching AI assistants how to perform complex tasks through well-structured instructions.

Skills

Claude-to-Kiro

Converts Claude Code configurations into equivalent AWS Kiro IDE/CLI configurations. Kiro adopted many of Claude Code's primitives (SKILL.md format, hook event model, MCP protocol), making most conversions near-direct.

Four conversion workflows:

• Single skill — one Claude skill → .kiro/skills/ (near-1:1, tool names translated)
• Full project — CLAUDE.md + skills + hooks + MCP → complete .kiro/ structure
• Global config — ~/.claude/ → ~/.kiro/ equivalents
• Plugin-to-power — plugin bundle → coordinated Kiro agents + steering + hooks + MCP

Key features:

• Signal-based content classification routes CLAUDE.md content to steering files with appropriate inclusion modes (always, fileMatch, manual, auto)
• Automatic tool name translation (Claude Read → Kiro read, Bash → shell, etc.) with mapping comment headers for traceability
• Five-section migration report including Enhancement Opportunities — Kiro features you could adopt that Claude Code doesn't have
• Plain-English migration guide for users new to Kiro covering rules, memory, security, skills, hooks, and agents

Quick start: "Port my Claude skills to Kiro" or "Convert this project's Claude setup to Kiro"

Documentation:

• Design document — full architecture, concept mapping, worked examples
• Design document (HTML) — styled version for easy reading
• Implementation plan

---

Claude-to-AmazonQ

Converts Claude Code configurations into Amazon Q Developer equivalents. Uses signal-based content classification to split Claude's unified config model across Amazon Q's component model (rules, prompts, custom agents, MCP config).

Quick start: "Port my deployment skill to Amazon Q" or "Convert this project's Claude setup to Amazon Q"

Documentation: Design document

---

Udemy Scanner

Searches and extracts structured course information from Udemy using multi-strategy fallbacks (Class Central mirrors, WebSearch aggregation). Supports single course lookups, multi-course comparisons, and topic-based bulk discovery.

Quick start: "Search Udemy for Python networking courses" or "Compare these two Udemy courses"

---

Repo Structure

Each skill follows a standard layout:

<skill-name>/
├── docs/
│   └── <skill-name>-design.md    # Design document (problem, architecture, examples)
└── skills/
    └── <skill-name>/
        ├── SKILL.md               # Skill definition (metadata, workflows, triggers)
        └── references/            # Templates and mapping guides for output formats

See INDEX.md for the complete skill catalog with function descriptions and example trigger phrases.

For New Users

These skills are designed for Claude Code (Anthropic's CLI). Install a skill by copying its skills/<skill-name>/ directory into your Claude Code skills location (~/.claude/skills/ for global, or a project's skills directory).

Each skill's SKILL.md contains trigger phrases in its YAML frontmatter — Claude automatically activates the skill when your request matches those phrases. No manual invocation needed.

Contributing

When adding a new skill:

1. Create a directory at the repo root following the standard structure above
2. Write a design document in docs/ before implementing the skill
3. Add a row to INDEX.md with the skill name, directory, function summary, and example trigger phrases

---

Created by: DT74 — Daren Threadingham