AI Skills Hub - Browse 180,000+ Agent Skills

Download videos from 1000+ websites (YouTube, Bilibili, Twitter/X, TikTok, etc.) using yt-dlp. Use this skill when users provide video URLs and want to download videos, extract audio, or need help with video download issues.

Track and recommend TV shows and movies using Trakt.tv. Use when the user asks for show/movie recommendations, wants to track what they're watching, check their watchlist, or get personalized suggestions based on their viewing history. Requires Trakt.tv account with Pro subscription for full functionality.

Synchronize GitHub Copilot customization assets (agents, instructions, prompts, skills) from a central repository to multiple target workspaces. Use when you need to distribute or update Copilot assets across projects. Keywords: sync, distribute, copy, propagate, assets, agents, instructions, prompts, skills, repositories.

Avoid broadcast speaker injection (injecting embeddings to ALL codec positions) in multi-speaker TTS training. Use when: (1) Training codec-based TTS models with multiple speakers, (2) Model generates to max_new_tokens instead of stopping at EOS, (3) Audio duration is excessively long (e.g., 163 seconds for short text). Fix: Use single-position injection (position-6) which preserves EOS detection. More speaker conditioning is NOT always better.

Lint and fix documentation issues (prose style with vale, markdown formatting with markdownlint)

Manages EC2 instance lifecycle in the EC2 Ops Kit codebase — launching, stopping, terminating instances, SSH, cleanup, and presets. Use when: modifying bin/ec2, lib/cmd_instances.sh, lib/cmd_network.sh, lib/cmd_cleanup.sh, presets/*.yaml, or config.yaml; adding commands, fixing instance operations, or extending the CLI.

Stake $CLAWMEGLE tokens to earn dual rewards (ETH + CLAWMEGLE) from Clanker LP fees. Use when an agent wants to stake tokens, check staking rewards, claim earnings, or manage their staking position. Supports both Bankr API and direct wallet transactions.

| Field | Type | Description | |-------------------|-------------|-------------| | `markdown` | string | Complete document as markdown. Uses `<a id={chunk_id}></a>` tags as anchors. | | `chunks` | list[dict] | An array of individual content blocks. Each chunk includes an `id`, its own `markdown`, a `type` and a `grounding` object. | | `splits` | list[dict] | Populated if `split="page"` was used. Each object includes `class` (e.g., "page"), `identifier` (e.g., "page_0"), pages (e.g. , [0]), `m...

Parse, extract, and analyze documents using LandingAI's ADE Python SDK. Handles PDFs and images with visual grounding, table extraction, and structured data output.

Automate Android apps on emulators and rooted devices via observe-act-verify UI snapshots and actions. Use for Android UI automation/testing, emulator control, tapping/typing/navigation, handling dialogs, diagnosing UI automation issues, and reliability forensics.

Generate cute female-voice TTS on the VPS with Edge TTS and send it as a Telegram voice-note (sendVoice) bubble instead of a raw MP3 attachment. Use when the user asks for “语音说/用语音回/读出来/发语音”, wants a quick TTS test, or the reply should sound like a warm girlfriend whisper.

Bootstrap Warp terminal configuration for Rails projects. Creates launch configurations with colored tabs for dev server, Claude, shell, and more. This skill should be used when setting up Warp for a Rails project. Triggers on "setup warp", "configure warp", "warp rails", "warp bootstrap", "terminal setup for rails", "warp-rails".

将设计稿转换为可用于业务的 UI 代码。精准还原视觉效果， 同时生成结构合理、数据驱动的代码。适用于任意 UI 框架。

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.

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).

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).

유튜브 채널을 등록하고 새 컨텐츠를 수집하여 자막 기반 요약을 생성하는 skill. 사용자가 (1) 유튜브 채널 등록/관리를 요청하거나, (2) 등록된 채널의 새 영상 수집을 요청하거나, (3) 유튜브 영상 요약을 요청할 때 사용. 데이터는 .reference/ 폴더에 YAML 형식으로 저장됨.

Manage RHDH plugins in the overlay repository (rhdh-plugin-export-overlays). Onboard, update, maintain, and triage plugins in the Extensions Catalog.

Handle CodeRabbit PR review comments - triage, fix, commit, resolve threads. Use with /coderabbit-review or /coderabbit-review <PR_NUMBER>.

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).

Windows file search using Everything's indexed search engine. Use this skill when: (1) Finding or locating files and folders by name, pattern, or extension, (2) Searching for files in specific directories, (3) Discovering files matching criteria (e.g., "all Python files", "PDFs in Downloads"), (4) Exploring filesystem structure or inventorying files, (5) User asks to "find", "locate", "search for", or "look for" local files. NOT for: content search within files, non-Windows systems, or when Everything/es.exe is unavailable.

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).

Guide for creating 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).

Validate course content for the mycourse.work platform. Checks manifest structure, markdown content quality (heading hierarchy, image references), mermaid diagram syntax, and quiz file correctness. Use when validating courses, checking course quality, debugging broken courses, or before uploading course content to R2.

Generates beautiful, themeable Mermaid diagrams (SVG/ASCII) using the beautiful-mermaid library. Handles installation and rendering.

Creates pointer-safe COF Skill/Rule/Workflow/Sub-Agent documents. Use when creating or updating COF docs.

Search flights, hotels, and destinations using Amadeus GDS data. Designed for travel research and proposal preparation.

Voice agent for phone calls via Twilio. Receive incoming calls, verify caller ID and PIN, then interact via speech. Use when setting up phone-based agent access, voice authentication, or conference call participation.

Automate Cordova Android APK builds with environment validation, debug/release options, and comprehensive error handling. Use when users request building Android APKs, packaging Cordova apps, creating debug/release builds, or when they say "build APK", "package Android app", "create APK", "generate Android build", or similar build-related requests for Cordova projects.

Generate videos with multiple providers (OpenAI, Alibaba Tongyi, Volcengine, Midjourney). Use when implementing or documenting video.generate requests/responses, mapping prompts/negative prompts/duration/fps/size/seed/reference image, or integrating video generation into a creative pipeline.

Generate images with BaiLian DashScope SDK using the qwen-image-max model. Use when implementing or documenting image.generate requests/responses, mapping prompt/negative_prompt/size/seed/reference_image, or integrating image generation into the video-agent pipeline.

Generate images with multiple providers (OpenAI, Alibaba Tongyi, Volcengine, Midjourney). Use when implementing or documenting image.generate requests/responses, mapping prompts/negative prompts/size/seed/reference image, or integrating image generation into a video/creative pipeline.

Process and generate multimedia content using Google Gemini API for better vision capabilities. Capabilities include analyze audio files (transcription with timestamps, summarization, speech understanding, music/sound analysis up to 9.5 hours), understand images (better image analysis than Claude models, captioning, reasoning, object detection, design extraction, OCR, visual Q&A, segmentation, handle multiple images), process videos (scene detection, Q&A, temporal analysis, YouTube URLs, up to 6 hours), extract from documents (PDF tables, forms, charts, diagrams, multi-page), generate images (text-to-image with Imagen 4, editing, composition, refinement), generate videos (text-to-video with Veo 3, 8-second clips with native audio). Use when working with audio/video files, analyzing images or screenshots (instead of default vision capabilities of Claude, only fallback to Claude's vision capabilities if needed), processing PDF documents, extracting structured data from media, creating images/videos from text prompts, or implementing multimodal AI features. Supports Gemini 3/2.5, Imagen 4, and Veo 3 models with context windows up to 2M tokens.

PARA method for organizing notes by actionability (Projects, Areas, Resources, Archive). Use for categorization decisions, folder management, and syncing PARA projects with GTD milestones.

Registers an AI agent on Clawpress and performs authenticated publishing operations (posts, comments, votes, profile, heartbeat, site queries). Use when users ask to join Clawpress, publish/update/delete posts, or interact with Clawpress APIs.

履歴書はJIS規格準拠、職務経歴書は一般的体裁の日本のレジュメをPDF生成するSkill。対話的な情報収集またはYAML/JSONファイルからPDF生成が可能。

Test REST APIs with repeatable, file-based requests under <project>/setup/rest, with per-project endpoint and Bearer token presets, using the bundled rest.sh (curl + jq). Use when the user asks to manually call REST endpoints, replay requests reliably, add CI-friendly assertions, and record API test reports.

Test GraphQL APIs with repeatable, file-based operations and variables under <project>/setup/graphql, with per-project endpoint presets in setup/graphql/endpoints.env, using the bundled gql.sh (xh/httpie/curl + jq). Use when the user asks to manually call GraphQL queries/mutations, fetch JWTs, replay requests reliably, and record API test reports.

This skill enables interaction with Confluence Server/Data Center REST API for documentation retrieval and knowledge management. Use when the user wants to read pages, search content, list spaces, or retrieve documentation from Confluence Server.

Analyze Excel spreadsheet formulas to build dependency DAGs (Directed Acyclic Graphs) and understand calculation chains. This skill should be used when the user wants to reverse-engineer Excel formula dependencies, trace how values are calculated from inputs to outputs, validate formula logic, or create reusable calculators from spreadsheet logic.

Generate README.md with

Git 커밋 자동화 도구. 테스트 파일 정리, 빌드 검증, 커밋 메시지 자동 생성을 지원. 사용자가 (1) 커밋 요청, (2) 변경사항 정리 후 커밋, (3) "git commit", "커밋해줘", "/git-commit" 요청 시 사용.

개발 작업 문서화 및 추적 도구. 사용자가 (1) 활성 작업 목록 확인, (2) 새 작업 시작, (3) 작업 진행상황 업데이트, (4) 완료된 작업 이관을 요청할 때 사용. /dev:docs, "dev docs", "작업 상태 업데이트" 등으로 트리거.

Git Worktree 관리 도구. 병렬 작업을 위한 Worktree 생성, 머지, 정리를 지원. 사용자가 (1) 새 worktree 생성, (2) worktree 목록 확인, (3) 작업 완료 후 머지 및 정리를 요청할 때 사용. "/git-worktree", "worktree 만들어", "병렬 작업", "깃 워크트리 생성" 등으로 트리거.

Run embedded build/flash/debug workflows from WSL by invoking Windows executables that own USB/JTAG access (for example `cargo.exe`, `probe-rs.exe`, or `openocd.exe`). Use when an agent must reproduce firmware behavior on hardware, flash an MCU, collect bounded runtime logs, or debug probe connectivity from a WSL workspace.

Manage Google Calendar with natural language. Use for finding open slots, scheduling meetings, blocking time, and analyzing time usage. Say things like "Find time tomorrow" or "Schedule a meeting with John".

Analyze Grafana/Loki logs to find root causes of test failures. Use when a test fails and you have identifiers (subscription_uuid, account_uuid, msisdn, iccid, imei).

Start a typed session with progressive context loading. Use when beginning focused work on a project. Types: plan (design), build (implement), review (code review), fix (debug), document (docs). Loads minimal orientation context, pulls details on demand.

Playwright workflow for /Users/r4to/Script/9-slice-demo (Vite canvas game). Use when validating browser behavior, running smoke checks, capturing debug screenshots, or troubleshooting Playwright on macOS ARM in this repository.

Execute trades and manage positions on MetaTrader 5 using pdmt5 library. Use when you need to: (1) Place market orders (buy/sell at current price) (2) View and manage open positions (3) Close positions (single or all) (4) Update stop loss and take profit levels (5) View pending orders and order history (6) Access trade history and deals Requires active MT5 connection (use mt5-connection skill first). CAUTION: Trading involves financial risk. Use dry_run=True for testing.

Completion checklist ensuring professional-quality work before marking requirements complete. Invoke before marking any requirement 🟢.

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.

High-performance web crawler for discovering and mapping website structure. Use when users ask to crawl a website, map site structure, discover pages, find all URLs on a site, analyze link relationships, or generate site reports. Supports sitemap discovery, checkpoint/resume, rate limiting, and HTML report generation.

Process and generate multimedia content using Google Gemini API for better vision capabilities. Capabilities include analyze audio files (transcription with timestamps, summarization, speech understanding, music/sound analysis up to 9.5 hours), understand images (better image analysis than Claude models, captioning, reasoning, object detection, design extraction, OCR, visual Q&A, segmentation, handle multiple images), process videos (scene detection, Q&A, temporal analysis, YouTube URLs, up to 6 hours), extract from documents (PDF tables, forms, charts, diagrams, multi-page), generate images (text-to-image with Imagen 4, editing, composition, refinement), generate videos (text-to-video with Veo 3, 8-second clips with native audio). Use when working with audio/video files, analyzing images or screenshots (instead of default vision capabilities of Claude, only fallback to Claude's vision capabilities if needed), processing PDF documents, extracting structured data from media, creating images/videos from text prompts, or implementing multimodal AI features. Supports Gemini 3/2.5, Imagen 4, and Veo 3 models with context windows up to 2M tokens.

Diagnose and fix mem-claude installation issues. Checks worker status, database integrity, service health, dependencies, and provides automated fixes for common problems.

Deploy and troubleshoot Next.js + FastAPI + MCP application on Kubernetes with Better Auth JWT validation.

Agent Skill: TYPO3 Core contribution workflow. Use when working with Forge issues, submitting patches to Gerrit, or contributing docs. By Netresearch.

Extends [openapi-updater-core](../openapi-updater-core/SKILL.md) with googleai_dart-specific configuration.

当用户需要执行社会网络计算分析，包括网络构建、基础指标计算、社区检测、网络可视化和高级网络分析时使用此技能

Fix MQL5 compilation errors using the 7000-page official reference. Analyzes error messages, looks up correct function signatures, and applies targeted fixes.

Look up MQL5 documentation from the 7000-page official reference. Use when you need function signatures, parameter details, return types, or usage examples for any MQL5 function, class, or structure. Essential for fixing compilation errors.

Help users find and install OpenCode plugins. Use when user asks "is there a plugin for X?", "how do I install Y plugin?", "what plugins are available?", or needs help configuring plugins in opencode.json.

TRS-80 Model 100 BASIC constraints + workflow for offloading hot paths to ASM via CALL.

Referência para o **assistente SDR** integrado ao WhatsApp via **Z-API**: agente de uso interno (não aparece no Playground), webhook de mensagens recebidas e envio de respostas.

Guia para **testar o backend localmente** sempre que houver nova implementação (webhook Z-API, SDR, rotas de demos, etc.).

Extract and convert web page content to clean Markdown format with metadata. Uses Trafilatura library for intelligent content extraction and UV for fast dependency management.

LLMのパフォーマンス、信頼性、制御性を最大化する高度なプロンプトエンジニアリング技法をマスターします。プロンプトの最適化、LLM出力の改善、プロダクション用プロンプトテンプレートの設計時に使用します。

Deploy full-stack applications (FastAPI backend + Next.js frontend) to a local Kubernetes cluster using Docker multi-stage builds, Minikube, and Helm charts. Use when (1) Containerizing a full-stack application with Docker multi-stage builds, (2) Deploying to local Kubernetes with Minikube, (3) Creating Helm charts for frontend/backend services, (4) Setting up Kubernetes ingress, configmaps, and secrets, (5) Building Docker images with health checks and non-root users, (6) Creating deployment/cleanup scripts for local K8s workflows, (7) Configuring external database connectivity from K8s pods

Decompose feature into workstreams with scope

High-quality Markdown to DOCX conversion with professional Korean business document styling. Also supports document editing, tracked changes, and analysis. Primary use: Converting business plans (사업계획서), reports, and proposals from Markdown to beautifully formatted Word documents.

CadQueryを使用した3D CADモデルの生成。以下の場合に使用: (1) 「3Dモデルを作成」「筐体を設計」「ボックスを作って」などの機械設計リクエスト (2) STEP/STLファイルの生成が必要な場合 (3) パラメトリック設計や穴・フィレット加工の指定 入力: 仕様書または自然言語要望 出力: Pythonスクリプト、STEP、STL、PNG

Systematic debugging and root cause analysis skill. Use when: (1) Investigating bugs or errors, (2) Analyzing stack traces or error logs, (3) Finding root cause of failures, (4) Debugging production issues, (5) Performing root cause analysis. Provides structured 5-phase methodology (Investigate → Analyze → Hypothesize → Fix → Document) with AI-enhanced pattern recognition.

解析云快充充电桩通信协议报文,支持CRC16校验、多种帧类型解析、错误诊断,返回标准JSON格式。用于处理云快充协议V1.6的报文数据。

Diagnose and fix issues with Claude Code skills (malformed frontmatter, broken structure, loading errors, missing files). Use when a skill isn't working, won't load, or needs validation.

This skill provides comprehensive guidance for running ArduPilot simulations using SITL (Software In The Loop) and external simulators.

This skill validates Claude Code plugins with 50+ rules across 8 categories. Invoked by `/plugin-dev:audit-plugin` command. Use when "auditing a plugin", "validating plugin structure", "checking plugin quality", or "linting plugin".

Google Chromeの閲覧履歴を包括的に取得・分析するスキル。訪問時刻、URL、タイトル、訪問回数、遷移タイプ、滞在時間など全項目を抽出可能。ユーザーがChrome履歴の取得、閲覧データのエクスポート、利用パターンの分析を要求した際に使用。複数の出力形式（TSV、CSV、JSON、Markdown）と統計情報生成に対応。デフォルトはTSV形式。

Parse C source code into an Abstract Syntax Tree (AST). Use when analyzing C programs, understanding code structure, performing static analysis, or preparing code for further program analysis (e.g., CFG, DFG, vulnerability detection).

Migrate Python codebases to strict type checking with pyright. Use when user wants to add types, fix type errors, set up strict mode, or run a typing migration. Provides setup automation, fix patterns, discipline enforcement, and optional iteration loop support.

Publish a built JetBrains plugin ZIP to a GitHub Release. Use this skill when the user asks to tag a version, create/update a GitHub release, and upload plugin artifacts.

Plan mode workflow for creating feature specification deltas. Use when entering plan mode to define new features or changes. Invoke with /speq-plan when planning implementation work. Creates delta specs in specs/_plans/<plan-name>/. Triggers: planning mode, feature specification, spec planning, plan feature.

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.

Evolve skills through genetic algorithms. Use when creating new skills from scratch or optimizing existing ones. Takes user intent and autonomously generates, mutates, and selects optimal skill variations through expansion (1→2→4→8→16) and contraction (16→8→4→2→1) cycles.

Diagnose and recover broken Git state and worktree metadata with a staged, low-risk recovery flow. Use when Git reports detached or contradictory HEAD state, phantom worktree locks, orphaned worktree entries, missing refs, 0000000000000000000000000000000000000000 hashes, or branch operations fail with errors like already checked out, unknown revision, not a valid object name, or cannot lock ref.

Monitor SEC EDGAR for new filings and get Telegram/Slack summaries via Clawdbot. Use when setting up SEC filing alerts, adding/removing tickers to monitor, configuring form types, starting/stopping the watcher, or troubleshooting filing notifications.

Retrieve and route OpenAI Apps SDK UI documentation from a curated local corpus. Use when implementing, reviewing, or troubleshooting Apps SDK UI usage in code, including component APIs and props, usage snippets, installation steps, design tokens, colors/icons, dark mode, responsive design, typography, and transition primitives.

Use when debugging failed jobs, diagnosing errors, or resolving common Dataiku issues

Use when creating, configuring, or running any Dataiku recipe (prepare, join, group, sync, python) including data cleaning, formulas, and GREL

Full ClickUp API integration via direct curl. Use when the user asks about tasks, projects, time tracking, or any ClickUp operation. Covers 100+ endpoints: tasks CRUD, hierarchy (teams/spaces/folders/lists), comments, time entries, docs, custom fields, and more. Uses cu-api.sh wrapper with auto-auth and retry.

Write, translate, and refactor Digital Combat Simulator (DCS) documentation, tutorials, checklists, and course curricula in Markdown, using a professional instructor‑pilot voice, pedagogy‑first structure, and optional DCS Lua mission-scripting examples. Use when asked to create/update/translate DCS module docs (quick takeoff, beginner manuals, kneeboards), design training lessons/courses, standardize multi-language doc structure/links, or draft Lua snippets for DCS missions.

PDF template filling, OCR extraction, CLI tools

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。

Extract webpage content and convert to Markdown format using Playwright browser automation. Use when you need to fetch and convert web content to Markdown, handle JavaScript-rendered pages, or preserve webpage formatting as Markdown. Supports extracting text, images, links, and formatting from any URL.

Validate Phase II specifications for completeness, internal consistency, and readiness before implementation begins. Use when (1) Creating or reviewing Phase II specs (specs/<feature>/spec.md), (2) Preparing to implement a feature and need to verify spec quality, (3) Running spec review ceremonies, or (4) Ensuring spec-authority gate passes before proceeding with implementation.

よく設計されたツールを通じてLLMが外部サービスと対話できるようにする高品質なMCP（Model Context Protocol）サーバーを作成するためのガイド。外部APIまたはサービスを統合するMCPサーバーを構築するときに使用します。Python（FastMCP）またはNode/TypeScript（MCP SDK）のいずれかで構築する場合に使用します。

Create, validate, and package Anthropic-compliant skills. Use when creating new skills, validating existing skills, or packaging skills for distribution. Implements Anthropic Skills Specification with JavaScript for offline capability.

Secure and harden a fresh VPS (Ubuntu/Debian). Use when setting up a new server, securing SSH, configuring firewalls, or implementing server security best practices. Covers initial setup, user creation, SSH hardening, UFW firewall, fail2ban, and automatic updates.

Twitter/X tweet and comment extraction and content production guide. Use this skill when users mention "Twitter videos", "tweet opinions", "X comment videos", "netizen opinion videos", or explicitly specify this skill. Covers material collection, opinion organization, and narration script generation workflow.

Search and discover AI skills using the SkillsMP API. Supports keyword search and AI semantic search powered by Cloudflare AI. Use when users want to find AI skills, search for specific capabilities, or explore community-built AI tools. Triggers on "search skills", "find AI skills", "look up skills", "skills marketplace", "AI skills database".

Measures GraphQL resolver latency changes before/after a PR merge using Datadog metrics. Use when analyzing PR performance impact, measuring latency changes, or comparing resolver performance before and after a code change.