AI Skills Hub - Browse 180,000+ Agent Skills

GMCC boot validation system. Automatically initializes GMCC environment on SessionStart. Invoke directly for boot diagnostics and troubleshooting.

Manage Notion-backed tasks via the ntask CLI. Use when creating tasks, checking task queues, claiming work, updating task status, adding comments, blocking or completing work items, or any project management operations backed by Notion. All task operations are performed exclusively by executing the ntask binary — never by calling the Notion API directly.

Real-time monitoring dashboard for PAI multi-agent activity. USE WHEN user says 'start observability', 'stop dashboard', 'restart observability', 'monitor agents', 'show agent activity', or needs to debug multi-agent workflows.

Search, read, create, and update Confluence pages. This skill should be used when working with company wiki, documentation, knowledge base content, or when users mention Confluence, wiki, or internal docs.

Interactively create and ingest a knowledge base from crawled URLs, single pages, or Firecrawl JSON folders with real-time progress and detailed statistics

Use when creating or populating a PR description from git changes. Triggers on requests to write PR summaries, fill out PR templates, or prepare PR descriptions before opening a pull request.

Intelligent vehicle control system for OpenClaw service supporting navigation, media control, climate systems, seat adjustments, lighting control, vehicle modes, and advanced scenarios. Use when users request vehicle control operations like "turn on AC", "play music", "navigate home", "enable nap mode", "adjust seat heating", "turn on reading lights", "set temperature to 25", "control ambient lighting", "adjust rear screen", or any combination of vehicle functions. Supports simple commands, complex multi-step operations, and scenario-based requests like "prepare for driving", "relax mode", "cool down the car", or "bedtime setup".

Send email through Proton Mail Bridge (localhost SMTP) using age-encrypted credentials. Use when setting up Proton Bridge for an agent mailbox, encrypting Bridge credentials (no 1Password), or sending automated emails (daily reports, alerts) via Proton Bridge.

使用 302ai-custom-mcp 生成图片并落地到项目 public 目录的完整流程。适用于用户要求“生成/创建图片、插图、配图、封面”或需要将生成图片下载到 public 并在代码中引用的场景。

将外部文件 URL 通过媒体上传接口转成站内可访问的文件 URL。适用于用户说“把外链文件转成内部链接/可访问链接/媒体 URL”、需要走 media.seedances.org 的 upload 接口、或需要生成 `https://media.seedances.org/media/download/{key}` 地址时。

Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.

gRPC 서비스 통합 가이드. Port/Adapter 패턴, Proto 정의, Async Client 구현 시 참조. "grpc", "proto", "subagent", "character", "location", "rpc" 키워드로 트리거.

Redis 기반 Composite Event Bus 패턴 가이드. Event Router, SSE Gateway, Redis Streams/Pub-Sub 구현 시 참조. "event", "sse", "stream", "pubsub", "broadcast", "realtime" 키워드로 트리거.

> Chat 서비스의 데이터 영속화 레이어 (Checkpointer + Message Storage)

Add Swift files to the AIQ Xcode project. Use when creating new Swift files (views, view models, services, models, tests) that need to be added to the Xcode project and build targets.

Run the question generation pipeline with optional type and difficulty filters. Generates questions, evaluates with judge, deduplicates, and inserts to database.

Add, remove, or query task dependencies in the local tasks database

Poll a GitHub PR for Claude's review comments. Waits for Claude to post a review and returns the content. Use after creating a PR to wait for automated review.

Browser automation, debugging, and performance analysis using Puppeteer CLI scripts. Use for automating browsers, taking screenshots, analyzing performance, monitoring network traffic, web scraping, form automation, and JavaScript debugging.

Validate UNS JSON structures against the standard three-tier hierarchy, type mapping, and alias uniqueness rules.

Use when working with PR review comments, resolving threads, or replying to discussion comments. Essential for understanding the correct erk exec commands for PR thread operations.

Fetches and classifies PR review feedback with context isolation. Returns structured JSON with thread IDs for deterministic resolution. Use when analyzing PR comments before addressing them.

Full-featured Salesforce development toolkit for deployment, testing, SOQL queries, org comparison, performance analysis, data migration, rollback, and health monitoring. Use for any Salesforce task including comparing environments, deploying components, running tests, finding duplicates, optimizing performance, error troubleshooting, or CI/CD integration.

Agent 채팅 데이터 무결성 관리 스킬. IndexedDB v3 스키마, Optimistic Updates, Eventual Consistency 패턴 구현 시 사용.

通过规范驱动的方法创建结构化的变更提案与规范差异。用于规划功能、创建提案、编写规范、引入新能力或启动开发流程。触发词包括 「规划」「创建提案」「规划变更」「规范功能」「新功能」「新特性」「新需求」「添加功能规划」「设计规范」。

Comprehensive market environment analysis and reporting tool. Analyzes global markets including US, European, Asian markets, forex, commodities, and economic indicators. Provides risk-on/risk-off assessment, sector analysis, and technical indicator interpretation. Triggers on keywords like market analysis, market environment, global markets, trading environment, market conditions, investment climate, market sentiment, forex analysis, stock market analysis, 相場環境, 市場分析, マーケット状況, 投資環境.

Create or update GitHub Pull Requests with the gh CLI, including deciding whether to create a new PR or only push based on existing PR merge status. Use when the user asks to open/create/edit a PR, generate a PR body/template, or says 'PRを出して/PR作成/gh pr'. Defaults: base=develop, head=current branch (same-branch only; never create/switch branches).

Inspect GitHub PR checks with gh, detect merge conflicts/update-branch requirements, list reviewer feedback, pull failing GitHub Actions logs, summarize failure context, then create a fix plan and implement after user approval. Use when a user asks to debug or fix failing PR CI/CD checks on GitHub Actions and wants a plan + code changes; for external checks (e.g., Buildkite), only report the details URL and mark them out of scope.

Port ESLint rules to rslint (a Go-based linter). Use when users want to port/migrate/implement an ESLint rule, add a new lint rule, or mention "port rule" or specific ESLint rule names like "no-unused-vars", "no-console", etc. This skill guides through the complete workflow from setup to PR submission.

Record and transcribe voice input when user wants to speak instead of type, describe complex issues verbally, provide audio input, or dictate text. Use this when user says "record my voice", "let me speak", "voice input", "transcribe audio", or when verbal description would be clearer than typing.

Detect and clean up zombie Playwright browser processes that cause endless tab loops during browser automation. Use this skill before invoking any browser-based agents (chatgpt-citation-checker, gemini-citation-checker) or when user reports stuck/multiplying browser tabs. Also useful for manual cleanup when browser automation fails or hangs.

Analyze large CSV files using efficient filtering and aggregation. Use when user asks about CSV data, trends, summaries, or top/bottom records.

Integrate Tauri backend with React frontend via IPC. Use when implementing command invocation from React, handling Tauri events in React components, streaming progress updates, managing async operations, syncing frontend-backend state, or building type-safe IPC layers.

建立與管理遵循 Anthropic 最佳實務的 Claude Code 技能。適用於建立新技能、修改 skill-rules.json、理解觸發模式、使用 hooks、除錯技能啟動問題，或實作漸進式揭露。涵蓋技能結構、YAML frontmatter、觸發類型（關鍵字、意圖模式、檔案路徑、內容模式）、執行層級（block、suggest、warn）、hook 機制（UserPromptSubmit、PreToolUse）、會話追蹤，以及 500 行規則。

Create, update, or review Standard Operating Procedures (SOPs), work instructions, process documentation, and procedural guides. Use when users request help writing SOPs, documenting processes, creating procedure manuals, standardizing workflows, or need guidance on operational documentation.

Calculate true disease burden per capita by combining WHO health indicators with Data Commons population data. Demonstrates multi-server integration with intelligent per-capita rate calculation. Handles both raw counts (e.g., death counts) and pre-calculated rates (e.g., mortality per 100k).

DO NOT USE THE APPLICATION NOTE SERVICE. Maintain and organize daily notes for the current project. Create, update, and manage these notes in a structured manner upon user request. Use this skill when the user requests "please make|create|write|add a note...".

RST/Markdown writing standards, Vale rules, YAML data management, and content guidelines for Write the Docs. Use when writing documentation, creating conference pages, or ensuring content meets style requirements.

Creates visually appealing MermaidJS diagrams with proper layout, alignment, and structure. Use when generating flowcharts, sequence diagrams, class diagrams, ER diagrams, state diagrams, Gantt charts, pie charts, mindmaps, or any MermaidJS visualization. Triggers on requests for diagrams, flowcharts, visual workflows, architecture diagrams, or when .mmd files are involved.

按照 Anthropic 最佳实践创建和管理 Claude Code 技能。在创建新技能、修改 skill-rules.json、理解触发模式、使用钩子、调试技能激活或实现渐进式披露时使用。涵盖技能结构、YAML 前言、触发类型（关键词、意图模式、文件路径、内容模式）、执行级别（block、suggest、warn）、钩子机制（UserPromptSubmit、PreToolUse）、会话跟踪和 500 行规则。

Search arXiv preprint repository for papers in physics, mathematics, computer science, quantitative biology, and related fields

Automate article validation, semantic commit generation, and git publishing for MDX documentation. Validates syntax, runs build checks, creates semantic commits, and pushes to remote repositories.

Автоматизация выполнения одной фазы из готового phase file с checkpoint validation и structured output

Code architecture reference for iclaude.sh (9 components + critical functions)

Best practices, common pitfalls, and development guidelines for iclaude.sh

Analyze and prioritize MultiQC module requests through automated triage. Calculate priority scores, assign labels, update project boards, and provide actionable feedback to contributors.

This skill should be used when the user asks to "run autoclaude", "automation", "one-shot", "single prompt", "batch automation", "loop automation", uses {autoclaude}, {oneshot}, {one-shot}, or needs Claude Code CLI automation (loop or one-shot).

Audit and improve AGENTS.md files in repositories.

Search arXiv preprint repository for papers in physics, mathematics, computer science, quantitative biology, and related fields

Control and test Flutter apps via semantics tree. Use when interacting with Flutter UI, automating forms, testing navigation, debugging UI state, or developing with hot reload. Always call list_elements first to discover available selectors.

Run verification commands and confirm output before claiming success

Replace arbitrary timeouts with condition polling for reliable async tests

使用 Playwright 与本地 Web 应用交互并进行测试的工具包，可验证前端功能、调试 UI 行为、捕获浏览器截图、查看日志。

Use CloudBase document database Web SDK to query, create, update, and delete data. Supports complex queries, pagination, aggregation, and geolocation queries.

Capture solved problems as categorized documentation with YAML frontmatter for fast lookup

환경(local, staging, production)별로 Docker Compose를 효과적으로 구성하는 방법을 다룹니다.

Land a feature branch: pull from main, rebase the branch, create a PR, and merge it via rebase with automatic branch deletion. Use when ready to land a completed feature branch.

Create branded PKG installers for plugin distribution. Use when user requests to package a plugin, create installer, share plugin with others, or mentions distributing/sending plugin to someone. Invoked by /package command or natural language like 'create installer for TapeAge' or 'package GainKnob'.

Ensures features have complete TanStack Query architecture (keys/options/hooks/queries)

// ❌ WRONG - hardcoded saveButton: page.getByTestId('blog-save-button'); ```

Creates ast-grep rules with validation and CI integration

Implement and monitor test pyramid distribution (unit/integration/e2e). Use when setting up test infrastructure, validating test categorization, or analyzing test health metrics. Supports Python (pytest), JavaScript (Vitest/Jest), Go, and Rust.

Search past reasoning for relevant decisions and approaches

Deterministic router for math cognitive stack - maps user intent to exact CLI commands

Render Mermaid diagrams as SVG or ASCII art using beautiful-mermaid. Use when users need to create flowcharts, sequence diagrams, state diagrams, class diagrams, or ER diagrams. Supports both graphical SVG output and terminal-friendly ASCII/Unicode output.

Oxygen Not Included production chain calculator with SQLite database extracted from decompiled game source

Modular workflow generator with on-demand context loading. Creates workflows from YAML plans with focused, trigger-specific guidance. Use when user wants to create, build, or generate workflow automation.

Download and analyze CBS (Dutch Statistics) Open Data datasets by ID. Use when user asks to analyze CBS datasets, explore energy/population/economic trends in Dutch data, query specific dataset IDs like 83140NED, or wants renewable energy trends, demographic analysis, or other Netherlands statistics.

Diagnostic and repair tool for dev-system. Checks hooks, settings files, path alignment, and common errors. Can auto-fix many issues.

Automatically renders DOT/Graphviz files to SVG and opens them in the browser. Provides instant visual feedback during Phase 2 Discovery when architecture diagrams are generated. Copyright (c) 2025 James J Ter Beest III. All Rights Reserved.

Automate Chinese novel creation with background research + multi-chapter drafting and Telegraph publishing, using the local `projects/novel-writer-cli` pipeline. Use when the user gives a story idea and wants you to (1) research and enrich background, then create a new novel project (topic/bible/characters/relations/8-chapter outline), (2) generate a specific chapter and publish/update it on Telegraph and return the URL, or (3) do basic novel operations like listing projects, switching the current project, checking status, and publishing the index page.

Manage secrets via GCP Secret Manager. Use when asked to set up secrets, migrate tokens to GCP, inject environment variables, or troubleshoot secret access issues.

Use when you need to research a target keyword before creating a page — discovers SERP competitors, analyzes their content strategy, builds a keyword pool, and outputs a structured Brief that feeds directly into /shipany-page-builder.

创建**全新动态页面**或优化现有页面的 SEO（主要针对英文或初始页面）。输入：关键词、路由/路径、参考内容。使用场景：(1) 从简短规格创建全新页面（关键词、路由、参考内容），(2) 增强现有页面内容以改善 SEO（添加章节、改进文案、扩展 FAQ），(3) 用户说'创建页面'、'优化页面 SEO'、'添加章节'，(4) 为新功能或产品创建落地页，(5) 创建英文源页面（后续可用 auto-i18n 国际化）。**不要用于：**(a) 国际化工作流（用 auto-i18n 创建或 i18n-content-audit 审计），(b) 审计现有i18n内容质量（用 i18n-content-audit），(c) 修复已有i18n页面的关键词问题（用 i18n-content-audit --fix），(d) 批量翻译页面（用 auto-i18n）。**核心功能：**创建新页面和内容优化，支持英文和非i18n场景，不处理国际化工作流。

Master error handling patterns across languages including exceptions, Result types, error propagation, and graceful degradation to build resilient applications. Use when implementing error handling, designing APIs, or improving application reliability.

Build web applications, dashboards, landing pages. Also handles UI revamps of existing codebases. Creates Modal dev servers for live preview with hot reload.

Transform prediction market data to Nixtla format (unique_id, ds, y). Use when preparing datasets for forecasting. Trigger with 'convert to Nixtla format' or 'schema mapping'.

Forecast multiple time series in parallel using TimeGPT. Use when processing 10-100+ contracts efficiently. Trigger with 'batch forecast' or 'parallel forecasting'.

Detects anomalies in time series data using TimeGPT. Identifies outliers, level shifts, and trend breaks without model training. Use when identifying anomalies, outliers, or unusual patterns in time series. Trigger with "detect anomalies", "find outliers", "anomaly detection".

Incorporates external variables (holidays, weather, events) into TimeGPT forecasts to improve accuracy. Use when forecasts require external data, holidays impact sales, or weather affects demand. Trigger with 'include holidays', 'add weather data', 'integrate events'.

Performs rigorous time series cross-validation using expanding and sliding windows. Use when needing to evaluate the performance of time series models on unseen data. Trigger with cross validate time series, evaluate forecasting model, time series backtesting.

Enables TimeGPT model fine-tuning on custom datasets with Nixtla SDK. Guides dataset preparation, job submission, status monitoring, model comparison, and accuracy benchmarking. Activates when user needs TimeGPT fine-tuning, custom model training, domain-specific optimization, or zero-shot vs fine-tuned comparison.

Turborepo monorepo build system guidance. Triggers on: turbo.json, task pipelines, dependsOn, caching, remote cache, the "turbo" CLI, --filter, --affected, CI optimization, environment variables, internal packages, monorepo structure/best practices, and boundaries. Use when user: configures tasks/workflows/pipelines, creates packages, sets up monorepo, shares code between apps, runs changed/affected packages, debugs cache, or has apps/packages directories.

Transforms forecasting experiments into production-ready inference pipelines with Airflow, Prefect, or cron orchestration. Generates ETL tasks, monitoring, error handling, and deployment configs. Activates when user needs to deploy forecasts to production, schedule batch inference, operationalize models, or create production pipelines.

Generate production-ready MCP server implementations from PRD tool specifications with schema validation, error handling, and testing infrastructure. Use when building MCP servers for Nixtla plugins, implementing tool handlers, or scaffolding server infrastructure. Trigger with 'build MCP server', 'generate MCP implementation', or 'scaffold MCP tools'.

Analyzes data sources and generates Nixtla-compatible schema transformations. Infers column mappings, creates transformation modules for CSV/SQL/Parquet/dbt sources, generates schema contracts, and validates data quality. Activates when user needs data transformation, schema mapping, column inference, or Nixtla format conversion.

Scaffolds production-ready forecasting experiments with Nixtla libraries. Creates configuration files, experiment harnesses, multi-model comparisons, and cross-validation workflows for StatsForecast, MLForecast, and TimeGPT. Activates when user needs experiment setup, forecasting pipeline creation, model benchmarking, or multi-model comparison framework.

Generate production-ready plugin structures from PRD documents with enterprise-compliant files. Use when scaffolding new plugins, converting PRDs to plugin skeletons, or initializing plugin projects. Trigger with 'scaffold plugin', 'create plugin from PRD', or 'initialize plugin structure'.

OWASP dependency vulnerability scan - Use OWASP Dependency-Check to detect known CVE vulnerabilities in project dependencies

Multi-language dependency security scan - Use Safety CLI and OSV-Scanner to quickly detect dependency vulnerabilities in Python/JS/Java projects

AI Agent code quality check - Use Ruff to check code standards for LangChain, AutoGen, and other AI Agent projects

Run multiple commands or tests in parallel with proper signal handling and exit code propagation.

Use CodexBar CLI local cost usage to summarize per-model usage for Codex or Claude, including the current (most recent) model or a full model breakdown. Trigger when asked for model-level usage/cost data from codexbar, or when you need a scriptable per-model summary from codexbar cost JSON.

Schedule and manage automated cron jobs on the Suruf platform. Create recurring tasks that run prompts on a schedule and post results to channels. Use when the user mentions scheduling, cron jobs, automated tasks, recurring jobs, or timed execution.

Download websites and web pages as clean Markdown files using the firemd CLI tool (powered by a local Firecrawl instance via Docker). Use when the user wants to: (1) Download a webpage or website as Markdown, (2) Scrape one or more URLs and save as .md files, (3) Batch download a list of URLs to Markdown, (4) Convert web content to Markdown for research, RAG, or archival, (5) Save web documentation locally as Markdown. Triggers on requests involving downloading sites, scraping URLs to Markdown, converting webpages to .md, fetching web content as text, or any mention of firemd.

Generate and edit images using Gemini via Vercel AI Gateway. Use when asked to generate images, create app icons, make hero images, edit photos, create OG images, generate product shots, or batch generate multiple images. Supports text-to-image, image editing with references, and batch processing.

Evaluate gen-blog output quality and regression risk by measuring dist size, posts index size, home payload, and cover image sizes. Use when reviewing generated static output (dist/) for performance or content degradation, especially after adding many posts or assets.

Robust URL-to-Markdown extraction for OpenClaw workflows. Use when the user wants to "extract/summarize/convert a webpage to markdown" (especially WeChat mp.weixin.qq.com) and web_fetch/browser is blocked or messy. Uses a cheap probe via web_fetch first, then falls back to the official MinerU API (via the local mineru-extract skill) and returns a traceable result contract with source links.

A gen-2 builder combining structured contracts with demo-first design for skills that are both agent-parseable and instantly tryable.

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes