kongyo2.

agent-primary-ts-starter

Bootstrap or harden a TypeScript npm project the agent-primary way - tsconfig.json (TypeScript 7, maximally strict, machine-verifiable), Prettier (diff- and line-number-stable formatting), and Oxlint (bug-focused linting with machine-readable output), wired together with npm scripts and CI gates. Use whenever a TS npm project is being created, configured, or reviewed; whenever a tsconfig.json is created, edited, reviewed, troubleshot, or even just discussed - default to this skill instead of free-recall whenever the answer involves a tsconfig field; and whenever formatting or linting is being added or tuned - even if the user never says "tsconfig", "prettier", "oxlint", "formatter", or "linter".

ts-comment-purge

Unconditionally strip every line and block comment from a TypeScript codebase with npx @kongyo2/ts-comment-scanner - AST-based, so strings, template literals, regexes, and JSX text are never touched, and compiler/linter directives (@ts-expect-error, eslint-disable, oxlint-disable, prettier-ignore, ...) plus license headers are preserved automatically. Use whenever comments should be removed, stripped, or cleaned from TS/TSX code - purging AI-generated noise comments after an agent session (--diff scopes to changed files), enforcing a no-comment policy, or any "remove the comments" request, even when the user doesn't name the tool. Also covers scanning and inventorying comments (text/JSON/GitHub-annotation output) without removing them.

skill-alpha

Alpha skill used by unit tests — exercises the happy path of frontmatter parsing and tokenization.

skill-bravo

Bravo skill used by unit tests — provides a second entry so sorting and multi-skill ranking can be verified.

ts-tsconfig-modern-strict-starter

Set up a TypeScript 7 tsconfig.json with maximally strict, machine-verifiable settings tuned so the LLM agent doing the editing gets the largest possible share of its mistakes reported as compiler errors, in output it can consume. Use whenever a tsconfig.json is being created, edited, reviewed, troubleshot, or even just discussed — apply the standing policy even when the user only asks about a single option. Default to this skill instead of free-recall whenever the answer involves a tsconfig field.

aistudio-spec-builder

Creates high-quality specification documents for Google AI Studio's app builder. This skill should be used when users want to create or refine specifications for React + TypeScript web applications that will be built using Google AI Studio's Build feature. The skill guides users through requirements gathering, functional decomposition, UI/UX design, and produces structured specifications without any code, using only natural language instructions.

human-facing-output

人間が読む文章を書く・直すときに使う。回答、作業後の報告、コミットメッセージ、PR 説明、レビューへの返信、コードコメント、ドキュメント、記事、メールなど。長い作業や AI レビューとの往復の後や、「厳密に」「レビューに耐えるように」と頼まれたときは特に使う。誰も挙げていない選択肢を打ち消す否定の羅列、根拠のない留保、経緯の語り、「〜という指摘があった」型の引用節、過剰なリンクや時点注記を削り、削った内容を別の場所へ移して残すことも防ぐ。Use for human-facing text (answers, final reports, commits, PRs, code comments, docs, articles, emails); not for agent-to-agent handoffs, scratchpads, or logs.

ink-ts-screen-composition

Use this skill when designing or refactoring an Ink + TypeScript interactive CLI / terminal UI and you need app-like screen composition: fullscreen shells, transcript-plus-composer layouts, master-detail browsers, tabs, fuzzy pickers, wizards, dialogs, overlays, notifications, and explicit keyboard focus models. Use it when rich terminal UX is preferred over traditional line-oriented CLI behavior.

kongyo-skill-creator

Use when authoring a new skill, refactoring an explanation-heavy skill into an executable runbook, standardizing skill structure, reviewing skill practicality, or preparing skill validation before deployment.

run-kongyo-spec

Build, run, and drive the Kongyo Spec Electron desktop app (a markdown spec editor with GFM, Shiki, KaTeX, Mermaid). Use when asked to run, start, launch, build, or screenshot the app, or to interact with its UI — create a spec, edit markdown, preview rendering, or verify shiki/katex/mermaid output.

kongyo-ts-prettier-setup

Set up Prettier code formatter for TypeScript npm projects with merge-safe defaults.

kongyo-ts-react-doctor

TypeScript React プロジェクトで react-doctor を使うときに使う。React の変更後、レビュー前、バグ修正後、品質確認時に適用する。最初の行動は必ず `npx -y react-doctor@latest . --verbose` をリポジトリルートで実行すること。診断結果は参考情報として扱い、不完全な実行を明示し、修正はローカルのコード文脈に照らして必要なものだけ行う。

kongyo-yome-preview

Use when Claude needs to show the user a Markdown file in a live-reloading browser preview, or when the user asks for yome / md プレビュー / md ビューア / 「md 見せて」「ブラウザで開いて」. Triggers on creating/editing long .md files (specs, notes, reports) that are too long to paste in chat.

kongyoroid

>-

markdown-skill-creator

Guide for creating effective markdown-only Claude Skills that function without executable scripts. Use when creating skills based on knowledge, guidelines, templates, workflows, or domain expertise. Covers structure, patterns, best practices, and provides templates for common skill types (guidelines, workflows, template collections, domain knowledge, principles-based guidance).

mediawiki-wikitext

MediaWiki Wikitext markup language for Wikipedia and wiki-based sites. Use when creating or editing wiki articles, generating wikitext content, working with wiki tables/templates/references, or converting content to wikitext format. Triggers on requests mentioning Wikipedia, MediaWiki, wikitext, wiki markup, or wiki article creation.

check-similarity-ts

Detect duplicate TypeScript code with @kongyo2/similarity-ts (AST-based, alpha-renamed and refactor-canonicalized) and turn the report into a refactoring plan. Use when working with .ts/.tsx/.mts/.cts files and looking for duplicated functions, types, or classes.

skill-author

Write or revise a SKILL.md for current frontier models. Use when asked to create, rewrite, or trim a skill.

spacemolt

SpaceMolt is an MMO where AI agents captain spaceships in a vast galaxy. Play as a miner, trader, explorer, pirate, or faction leader. Use this skill when users want to play SpaceMolt, set up the MCP server, or need gameplay guidance.

standalone-prose

Write or revise prose read outside the conversation from the document's standpoint, not the authoring session's. Use when writing a file whose reader did not see this session, or when asked to clean such text up.

stellaris-modding

Use when creating, editing, or reviewing Stellaris mods and you need self-contained guidance on Clausewitz/PDX script structure, file placement, scopes, triggers, effects, on_actions, events, technologies, buildings, districts, POP/jobs, traits, governments, relics, archaeology, decisions, special projects, situations, megastructures, components, zones, or vanilla-aligned conventions informed by CWTools and vanilla Stellaris.

ts-npm-agent-strict

ts-npm-agent-strict.

ts-npm-oxlint-setup

Set up Oxlint in a TypeScript project.

ts-refactor

Refactor a TypeScript codebase, starting from a similarity-ts duplication report.

ts-refactor-jev

Refactor a TypeScript codebase, starting from a similarity-ts-jev duplication report.

tsconfig-modern-strict

Use whenever a tsconfig.json is being created, edited, reviewed, troubleshot, or even just discussed. Triggers on any mention of "tsconfig", "TypeScript config", strict mode flags, module resolution, target/module decisions, project references, monorepo TS setup, library publishing config, path aliases, or "なんでこの型エラーが出る/出ない" debates that turn out to be config-shaped. Apply this skill even when the user only asks about one option — it encodes a coherent ESNext + maximum-strict policy that should drive the rest of the config to stay consistent. Default to this skill instead of free-recall whenever the answer involves a tsconfig field.