AI Skills Hub - Browse 180,000+ Agent Skills

한국근대사료DB(db.history.go.kr/modern)의 근대 사료를 검색어 또는 컬렉션 지정으로 크롤링하여 구조화된 JSON으로 저장. 원문과 번역문을 함께 수집. "크롤링" + "근대사료/근대/모던" 키워드 조합 시 사용.

Find and search for academic papers using the paper-finder service. Use when conducting literature review, searching for related work, finding baseline papers, or looking for methodology references.

Manage BibTeX citations including DOI-to-BibTeX conversion, metadata extraction, validation, and deduplication. Use when building reference lists, converting DOIs to citations, or validating existing .bib files.

This skill reviews all skills within a Claude Code plugin against Anthropic's official best practices to ensure they follow guidelines for conciseness, structure, progressive disclosure, and code quality.

Scrape webpage content, extract metadata, download images, and prepare for import/migration to AEM Edge Delivery Services. Returns analysis JSON with paths, metadata, cleaned HTML, and local images.

Execute R code through RStudio via MCP. Use for R programming tasks, statistical analysis, data visualization, or setting up/troubleshooting the RStudio integration with Claude Code.

つくよみちゃんTTS - 感情付きメッセージ読み上げ

Diagnose and self-heal an OpenClaw installation on Windows (gateway/service flapping, Feishu disconnects, browser CDP/headless issues, model auth 401/403/timeouts, Git conflicts/submodules, and GitHub push failures). Use when OpenClaw shows “Connection error”, frequent watchdog restarts, tool/API 401/404/429, “Invalid input[*].name … ^[a-zA-Z0-9_-]+$”, or when `git push` hangs/fails in `C:\Users\Administrator\.openclaw\workspace`.

Build UIs using shadcn/ui components from https://ui.shadcn.com. Use when the task mentions shadcn/ui, ui.shadcn.com, Radix-based UI components, `npx shadcn@latest` (create/init/add/mcp), `components.json`, shadcn registry items, or when you need to compose accessible React UI quickly with Tailwind + shadcn components.

Maintain and extend the win4r/AISuperDomain (AI超元域/Aila) desktop app configuration, especially config.json AiConfig entries that define embedded web-AI tabs (name/url/script/iniScript). Use when adding a new AI provider, fixing a broken automation script after a site UI change, validating config.json structure/IDs/placeholders, or extracting/updating the JavaScript DOM automation snippets.

Browser automation using Playwright MCP. Navigate websites, fill forms, click elements, take screenshots, and extract data. Use when tasks require web browsing, form submission, web scraping, UI testing, or any browser interaction.

标准作业程序（SOP）生成与规范化：将输入资料/需求/历史记录整理为可执行 SOP（结构化章节、步骤、控制点、异常处理、记录）。当用户要求“写 SOP/作业指导书/操作规程/流程说明”，或给出零散资料需要“整理成 SOP/流程”，或要求“按标准结构输出 SOP/质量检查”时使用。

Deploy Apache DevLake to Azure or run locally with Docker. Supports Official release images or Custom builds from source. Use when deploying DevLake, creating Azure resources for DevLake, setting up local DevLake, or troubleshooting DevLake deployments. Trigger phrases include "deploy devlake", "set up devlake", "devlake azure", "devlake local", "install devlake".

macOS-only AppleScript control for the ChatGPT Atlas desktop app. Use only when the user explicitly asks to control Atlas tabs/bookmarks/history on macOS and the "ChatGPT Atlas" app is installed; do not trigger for general browser tasks or non-macOS environments.

Create high-quality skills: scoped, procedural, and durable. Prefer updates over duplicates. Use when working with SKILL.md files, authoring new skills, improving existing skills, or understanding skill structure and best practices.

Fetches official documentation for external libraries and frameworks (React, Next.js, Prisma, FastAPI, Express, Tailwind, MongoDB, etc.) with 60-90% token savings via content-type filtering. Use this skill when implementing features using library APIs, debugging library-specific errors, troubleshooting configuration issues, installing or setting up frameworks, integrating third-party packages, upgrading between library versions, or looking up correct API patterns and best practices. Triggers automatically during coding work - fetch docs before writing library code to get correct patterns, not after guessing wrong.

Send emails via SMTP with support for plain text, HTML, and attachments. Use when the user asks to send an email, email someone, or compose and send a message. Supports single recipients and can include file attachments. Works with Gmail, Outlook, Yahoo, QQ Mail, 163 Mail, and any SMTP server.

Create Model Context Protocol (MCP) servers that expose tools, resources, and prompts to Codex. Use when building custom integrations, APIs, data sources, or any server that Codex should interact with via the MCP protocol. Supports both TypeScript and Python implementations.

Perform due diligence on crypto assets using the Messari API. Retrieves market data, tokenomics, funding history, upcoming unlocks, recent events, and research reports.

Use when the task requires automating a real browser from the terminal (navigation, form filling, snapshots, screenshots, data extraction, UI-flow debugging) via `playwright-cli` or the bundled wrapper script.

Go back to before the last task. Reverts the most recent change Director made.

Use when the task requires automating a real browser from the terminal (navigation, form filling, snapshots, screenshots, data extraction, UI-flow debugging) via `playwright-cli` or the bundled wrapper script.

Implement checks, patterns, and fixtures for an existing recommended rule skeleton

Fetch any URL (including JS-heavy or login-gated pages) and convert to clean markdown. Uses Chrome CDP for high-fidelity rendering.

Manages personal diary entries and performs cognitive audits (weekly, monthly, annual) using structured prompts. Use when the user asks to update their daily log, or requests a weekly, monthly, or annual review/audit of their diary.

Batch downloads images from standard repositories (tuanbiao) and automatically merges them into a PDF. Supports ID extraction from URLs. Use when downloading "团体标准" (tuanbiao) or standards from ttbz.org.cn.

Transform Markdown content into professional slide decks (PPTX/PDF). Features "Smart Resume" and "Editable Text".

Fetch and summarize YouTube video transcripts. Use when asked to summarize, transcribe, or extract content from YouTube videos. Handles transcript fetching via residential IP proxy to bypass YouTube's cloud IP blocks.

Secure SSH operations for Alibaba Cloud ECS instances through local SSH aliases and key-based authentication. Use when Codex needs to connect to Aliyun servers, run remote shell commands, inspect logs, transfer files with scp, or execute deployment and maintenance steps without exposing private keys or passwords.

TickTick/滴答清单统一管理 CLI,支持任务、项目、标签、评论和习惯管理。当用户需要以下操作时使用: (1) 创建/更新/删除/完成任务, (2) 管理项目, (3) 整理标签, (4) 添加任务评论, (5) 追踪习惯。所有操作需配置 `.env` 文件中的 DIDA_USERNAME 和 DIDA_PASSWORD。

文章简介，1-2句话概括

数据库 MCP 服务管理技能。当用户需要查询数据库、管理数据库配置、或使用 MCP 数据库功能时使用此技能。自动检测并安装 MCP 服务，提供便捷的数据库查询接口。

git worktreeを使用した並列実装オーケストレーション。「並列実装」「worktree agents」「タスク分割してブランチで実装」「仕様書を並列で実装」「複数機能を同時に開発」などのリクエスト時、または仕様書・タスクリストから複数の独立タスクを同時実装する場合に使用。

Access and manage the COMSA Dashboard PostgreSQL database. Use this skill when working with the comsa_dashboard database for queries, schema management, or data operations. Handles SSH tunnel for local development and assumes direct access on deployed servers.

Comprehensive spreadsheet creation, editing, and analysis with support for formulas, formatting, data analysis, and visualization

Use this skill when the user wants to generate a commit message. Contains workflow for branch checking, ticket ID extraction, and message formatting.

Patterns for finding and exporting mail via Microsoft Graph (delegated auth), with safe defaults and paging/retry guidance.

Download best-quality YouTube video/audio, burn in Korean subtitles with optional English subtitles above (dual-subtitle mode), then upload via YouTube Data API with Korean-translated title/description and unlisted privacy. Use when a user wants a hard-subbed version uploaded to a YouTube archive account.

Download YouTube audio with yt-dlp, transcribe with Whisper to create timestamped subtitles (SRT), then analyze the transcript to recommend 15–60s Shorts-ready highlight segments with timecodes, hook text, clear reasons, and editing/narration suggestions. Use when a user provides a YouTube link and asks for transcription/subtitles, highlight extraction, or Shorts clip recommendations.

Expert system for triaging CockroachDB roachtest failures

Use this skill when building MCP (Model Context Protocol) servers. This includes creating new MCP servers in TypeScript or Python, implementing tools with proper schemas, setting up transports, and following MCP best practices for tool design, error handling, and documentation.

Integrate Evernote note management into your development workflow. Use when you need to: (1) Search and retrieve notes by tag, title, or content, (2) Create new notes with proper tagging and notebook assignment, (3) Update existing notes (append content, add to sections, modify text), (4) Manage sprint/project notes with Tasks/Links/Notes structure, (5) List and manage tags across your Evernote account. Requires EVERNOTE_TOKEN environment variable set to your developer token.

Weekly SEO performance review for clients. Fetches Search Console data, compares week-over-week metrics, identifies top pages, rising/declining keywords, CTR opportunities, and generates a markdown report. Use when performing client SEO reviews or analyzing search performance trends.

Process aviation fuel surcharge bills from Excel files. Supports both automatic detection and Claude-assisted modes. Detects headers, matches columns, parses dates/routes, and fetches contract numbers via API. Use when working with aviation fuel bills, Excel file processing, or when user mentions fuel surcharges,航空燃油账单, or 燃油差价费.

Browser automation scripts for testing, screenshots, and web interaction. 16 scripts with persistent state.

Browser automation using Playwright MCP. Navigate websites, fill forms, click elements, take screenshots, and extract data. Use when tasks require web browsing, form submission, web scraping, UI testing, or any browser interaction. NOT when only fetching static content (use curl/wget instead).

UK police force identification via JavaScript pattern matching. ALWAYS use this before AI classification (see Guardrail G-005).

Comprehensive management tool for local skills in the skills directory. Use when Claude needs to: (1) List all available skills and their descriptions, (2) View detailed information about a specific skill (SKILL.md content and file structure), (3) Create new skills using templates, (4) Delete existing skills, (5) Package skills into .skill files for distribution, or (6) Search skills by name or description keywords.

Automated compliance evidence collection for AWS, GCP, and Azure with SOC 2, ISO 27001, NIST 800-53, and CIS benchmark mappings.

Post content to WeChat Official Account (微信公眾號). Supports both article posting (文章) and image-text posting (圖文).

Nemotron-Personas-Japan (1M件) から最適ペルソナを選定し、Instagram実データで検証・統合する全自動ペルソナ分析Skill。Instagram分析、ペルソナ作成依頼時に自動起動。

PDF-RAG API reference. REQUIRED after any failed curl/jq to localhost:8000 (404, null, jq error). Also use when uncertain about endpoint path or response shape.

Validates code quality before commits in the MigraineTracker React Native project. Use this skill when preparing to commit code, before pushing changes, or when the user asks to run pre-commit checks. Ensures TypeScript type safety, test coverage (80%+), passing tests, and proper branch strategy compliance.

各コンポーネントの詳細情報を構造化した「ドシエ（調査書）」を作成するスキル。 system-map/components.yamlを入力として、RAG検索の核となるコンポーネントカードを生成する。 LLMにコード全文を渡す代わりに、構造化されたカードで本質を伝える。

This skill should be used when the user asks to "split PDF by chapters", "divide book into chapters", "extract chapters from PDF", or "break PDF into sections". Splits PDF documents based on table of contents or text patterns using pypdf.

Format code using dotnet format, prettier, and other formatting tools. Use when task involves code style fixes, formatting consistency, or preparing code for commit.

Run static analysis, security scans, and dependency checks on .NET code. Use when task involves code quality, security audits, or vulnerability detection.

Build and orchestrate tasks using Nuke build system. Use when task involves running Nuke build targets, orchestrating complex build workflows, or managing build pipelines.

Interact with Dagster data orchestration platform running locally or on Kubernetes. Use when Claude needs to monitor Dagster runs, get run logs, list assets/jobs, materialize assets, launch jobs, or debug pipeline failures. Supports both local Dagster dev server and Kubernetes deployments where each job run is a separate pod.

This skill should be used when users need to fetch Jira ticket content using either a specific ticket ID (like RD-3891) or search for tickets by title/description. It defaults to searching within the current sprint but can extend to all tickets when needed. The skill uses the Jira CLI to retrieve ticket details, status, assignee, and descriptions.

Create checkpoints from receipt photos using QR scanning, e-Kasa API, and GPS extraction (10-40s)

Flutter development for screens, features, and business logic. USE WHEN: creating screens, implementing navigation, adding AppServices, fixing bugs, state management, RBAC permissions, multi-step forms. NOT FOR: single UI components <300 lines (use flutter-ui-components).

Gitコミット前に自動で品質チェックを実行するスキル。 「この修正をコミットして」「コードレビューしてからコミットして」などの依頼に対応し、 ESLint/BlackによるLint、テスト実行、コードレビューを自動実行後、 Conventional Commits準拠のコミットメッセージでコミットを作成する。 プッシュやPR作成は行わず、ローカルコミットまでを実行する。

Bundle size analysis, budget enforcement, and regression detection for Vite/Rollup builds. This skill should be used when: (1) analyzing bundle composition or total size, (2) checking size budgets, (3) comparing PR vs baseline for regressions, (4) investigating why bundle size increased, or (5) optimizing bundle size. Reads Vite manifest for chunk graph (Tier B). Cannot attribute sizes to specific npm packages without stats.json (Tier C unavailable by default).

Query real-time parking availability in Taipei City. Shows available parking spaces, capacity, status (green/available, orange/moderate, red/full), address, phone, and payment info. Supports underground/multi-level parking and street parking for cars, motorcycles, and buses. Use when user asks about parking availability, finding parking near a location, or checking parking occupancy in Taipei.

Empirical validation workflow for blockchain data collection pipelines before production implementation. Use when validating data sources, testing DuckDB integration, building POC collectors, or verifying complete fetch-to-storage pipelines for blockchain data.

Systematic workflow for researching and validating blockchain RPC providers. Use when evaluating RPC providers for historical data collection, rate limits, archive access, compute unit costs, or timeline estimation for large-scale blockchain data backfills.

Generates professional PDFs from Markdown using Pandoc with XeLaTeX. Use when creating PDFs, needing section numbering, table of contents, bibliography, or custom LaTeX styling.

Analyzes codebase to find similar features, reusable utilities, and architectural patterns

This skill generates planning documents (기획서) in PowerPoint format from analyzed screen data and project artifacts. It should be used by the auto-draft-orchestrator agent during Phase 4 to create the final PPT output. Not intended for direct user invocation.

Systematic code quality improvement workflow combining smart-reviewer and test-generator MCP tools with Pareto 80/20 auto-fixes, severity-based review, and comprehensive test generation. Use when p...

Specialized knowledge of Bash and Zsh scripting, shell automation, command-line tools, and scripting best practices. Use when the user needs to write, debug, or optimize shell scripts, work with command-line tools, automate tasks with bash/zsh, or asks for shell script help.

Orchestrate publishing monorepo packages to npm using Lerna with OTP handling, pre-publish validation (build + lint), dry-run preview, and post-publish verification. Use when user requests to publish, release, or deploy packages to npm registry. Designed for Lerna-managed monorepos with 2FA-enabled npm accounts.

LLM이 잘 설계된 도구를 통해 외부 서비스와 상호작용할 수 있게 하는 고품질 MCP(Model Context Protocol) 서버 생성 가이드. Python(FastMCP)이나 Node/TypeScript(MCP SDK)로 외부 API나 서비스를 통합하는 MCP 서버를 구축할 때 사용한다.

Add, review, or update GA4 event tracking on HTML websites using data attributes and a reusable JavaScript tracker. Use when wiring gtag or GTM dataLayer events, choosing event names/parameters, or extending a base GA4 tracking snippet for clicks and simple interactions.

Convert any MCP server into a Claude Skill with 90% context savings. Use this skill when converting an MCP server to a skill to reduce context usage and improve performance.

Blended voice profile

Combine multiple voice profiles with weighted mixing to create hybrid voices. Use when relevant to the task.

Browser automation using Playwright MCP. Navigate websites, fill forms, click elements, take screenshots, and extract data. Use for web browsing, form submission, web scraping, or UI testing. NOT for static content (use curl/wget).

Upload files to the self-hosted Lsky image bed (https://lsky.xueguo.us/api/v1), fetch the returned URLs/Markdown snippets, and manage albums or strategies via the REST API. Use this skill whenever you need to programmatically push screenshots/assets to Lsky and return shareable links inside Codex.

查詢 macOS 內建字典（日文、中文、英文）。 使用時機：當用戶詢問日文單字的讀音、詞性、定義，英文詞彙的中日文翻譯，或中文詞彙的英文翻譯時。 關鍵字：查字典、怎麼唸、什麼意思、日文定義、詞性、用日文解釋、英翻日、英翻中、中翻英。

A skill for implementing robust API request validation in Python web frameworks like FastAPI using Pydantic. Covers Pydantic models, custom validators (email, password), field-level and cross-field validation, query/file validation, and structured error responses. Use when you need to validate incoming API requests.

Intent-based specification tests evaluated by LLM-as-judge. Use when the user asks to "create spec tests", "write intent tests", "TDD with intent", "natural language tests", or wants tests that capture WHY, not just WHAT. NOT pytest/jest/unittest - natural language specs Claude evaluates.

Run and evaluate the Skillset smoke test (aliases, sets, and skill invocation) and interpret its JSON reports. Use this when validating $skill/$set behavior or checking Claude Code/Codex headless runs.

Generate POSIX shell scripts following UNIX programming principles and a rigorous coding style. Use when: (1) creating or updating shell scripts; (2) when working on .sh files; (3) when working on files with #!/bin/sh shebang lines. Do not use when: (1) asked to create a Bash script.

QuantConnect optimization API and Phase 4 parameter tuning (project)

Drizzle ORMを使用したデータベース操作を支援。スキーマ変更、マイグレーション、シードデータ投入、DBリセットで使用。トリガー例：「DBを更新」「スキーマ変更」「テーブル追加」「カラム追加」「シード実行」「マイグレーション」

Implement localStorage persistence in Next.js 14 App Router with SSR-safe patterns. Use when storing data in localStorage, handling hydration errors, implementing state persistence, or creating type-safe storage utilities.

Guides creation of high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or Node/TypeScript (MCP SDK). Covers tool design, authentication, Docker deployment, and evaluation creation. NOT when consuming existing MCP servers (use the server directly).

Systematic methodology for debugging bugs, test failures, and unexpected behavior. Use when encountering any technical issue before proposing fixes. Covers root cause investigation, pattern analysis, hypothesis testing, and fix implementation. Use ESPECIALLY when under time pressure, "just one quick fix" seems obvious, or you've already tried multiple fixes. NOT for exploratory code reading.

Write clear, effective technical documentation following industry-proven patterns from exemplary projects and authoritative style guides, with built-in countermeasures for common LLM documentation issues

Claude 스킬을 생성하고 작성하는 것을 도와줍니다. 새로운 스킬 만들기, 스킬 작성 가이드, 베스트 프랙티스 적용 등의 키워드에 반응합니다.

GLOWマスタデータCSVの検証スキル。作成したCSVファイルがDB投入可能か、server/client実装と整合性があるかをテンプレート、DBスキーマと照合してチェックします。マスタデータ、CSV検証、バリデーション、チェックで使用します。

Browser automation with Playwright for Python. Recommended for visual testing. (project)

This skill should be used when working with IMPLEMENTATION_PLAN.md files, parsing markdown checklists, reading verification commands from AGENTS.md, or tracking task completion status. Essential for agents like Ralph that follow structured implementation plans.

Dify Workflow/Chatflow作成・編集・バリデーション・インポートを行うスキル群。「Dify作成」「ワークフロー作成」「チャットフロー作成」などの依頼時に使用。

既存の完成済みDifyフローを対象に、特定ノードや一部要素のみを改善します。会話変数・プロンプト・ノード名・説明・UI要素・レイアウトを最適化し、元ファイルを直接編集して上書きインポートします。「フロー改善」「ブラッシュアップ」「カスタマイズ」「調整」などの依頼時に使用します。

DifyからURL抽出・対話選択・一括エクスポートを実行します。単一アプリのエクスポートは08_dify_auto_import.mdのdify_exportを使用し、バッチ処理・URL抽出・フィルタ・対話選択が必要な場合はこのルールを使用します。dify_batch_export.pyでURL入力からapp_id抽出、全アプリ一括エクスポート、名前フィルタ、対話形式での複数選択が可能です。

HTTP API経由でDify CloudにYMLを自動インポート/エクスポートします（Playwrightは認証Cookie取得のみに使用）。インポート：初回は新規作成、2回目以降は自動で上書きインポート。app_idは flow_log/.{ファイル名}.app_id に自動保存され、次回から同じフローを更新します。--newで強制新規作成、--app-idで手動指定も可能。エクスポート：app_id指定でDifyからYMLをダウンロード、--include-secretで機密情報込みエクスポート可能。

Difyフローのテスト実行・結果検証・ブラッシュアップループを行います。テスト要件書に基づいてテストを実行し、結果をMarkdownにまとめ、問題があれば修正→再テストのループを回します。「テスト実行」「テスト」「動作確認」などの依頼時に使用します。