# Help

> List every skill and specialist agent in this Minecraft-plugin toolkit, grouped by phase, with the recommended end-to-end workflow. Use this when the user asks what skills/commands are available, what this toolkit can do, which skill to use next, how the pieces fit together, or just types /help. Reads the skills and agents live from disk — never invents them.

- Skill: `itamarb2010-jpg/help` (Agent Skill)
- Install (CLI): `npx skillmds@latest add itamarb2010-jpg/help`
- Raw SKILL.md: https://api.skillmd.com/api/skills/itamarb2010-jpg/help/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: itamarb2010-jpg (https://skillmd.com/u/itamarb2010-jpg)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/itamarb2010-jpg/help

---


# Toolkit help — skills & agents

Give the user a skimmable map of everything available, read **live from disk** (so it stays
accurate as skills are added), grouped by the phase of building a Minecraft plugin.

## Phase 1: Discover what exists

- `Glob` `.claude/skills/*/SKILL.md` (fallback `**/skills/*/SKILL.md`). For each, read only the
  frontmatter `name` and `description` — don't dump the body.
- `Glob` `.claude/agents/*.md` (fallback `**/agents/*.md`). For each present, read the frontmatter
  `name`/`description`. If the directory is empty or absent, simply say there are no specialist
  agents yet — don't fabricate any.
- Only list skills/agents that actually exist. If one of the expected names below is missing,
  omit it silently rather than advertising a skill the user can't run.

## Phase 2: Present skills grouped by phase

Print a compact list — one line per skill: **`/name`** — a one-line gist distilled from its
description (not the full sentence). Group under these headings, in this order:

- **Setup** — get the target stack and project skeleton in place:
  `/start`, `/setup-platform`, `/scaffold`
- **Implement** — add features to the plugin:
  `/add-command`, `/add-listener`, `/add-config`, `/add-gui`, `/add-storage`
- **Ship** — build, test, and review before release:
  `/build`, `/run-server`, `/plugin-review`
- **Meta** — `/help` (this) — and mention `/start` as the guided entry point for newcomers.

Place any skill found on disk that isn't in the lists above under a **Other** heading so nothing
is hidden.

## Phase 3: List the specialist agents

Under an **Agents** heading, briefly list each agent from `.claude/agents/` — `name` — one-line
purpose — and note they're spawned for focused sub-tasks (deeper research, specialized reviews)
rather than invoked as slash commands. If none exist, one line: "No specialist agents installed
yet."

## Phase 4: Show the workflow arc

End with the recommended end-to-end path so the pieces connect, e.g.:

```
new plugin ─▶ /setup-platform ─▶ /scaffold ─▶ implement (/add-command,
             /add-listener, /add-config, /add-gui, /add-storage) ─▶ /build
             ─▶ /run-server ─▶ /plugin-review ─▶ ship
```

- Not sure where to start? Point to **`/start`** (guided routing).
- Note that most skills read the target stack from `.mcplugin/config.yml`, so `/setup-platform`
  comes first; if a skill complains the config is missing, run it.

Keep the whole output short and scannable — headings, one line per item, no long paragraphs. This
skill only reads and reports; it never runs another skill or edits anything.