AI Skills Hub - Browse 180,000+ Agent Skills

When adding a new test flow to the QA Playwright project, follow this checklist.

When adding or extending test data for a flow, follow this structure.

API gateway for calling third-party APIs with managed auth. Use this skill when users want to interact with external services like Slack, HubSpot, Salesforce, Google Workspace, Stripe, Shopify, and more.

Web 应用测试工具 - 使用 Playwright 与本地 Web 应用交互和测试的工具包。 支持验证前端功能、调试 UI 行为、捕获浏览器截图和查看浏览器日志。 当用户需要：测试 Web 应用、自动化浏览器操作、截图、调试前端时触发此技能。

帮我分析内容数据

图像裁剪工具，支持保持画面中心的智能裁剪。使用场景：（1）微信公众号封面裁剪（2.35:1）（2）社交媒体正方形裁剪（1:1）（3）批量处理图片尺寸（4）各种社交媒体预设尺寸（微信、微博、Instagram等）。当用户需要"裁剪图片"、"调整图片尺寸"、"生成正方形图片"、"微信封面裁剪"时触发此skill。

Excel 数据分析工具 - 创建、编辑和分析电子表格，支持公式、格式化、数据分析和可视化。 当用户需要：创建 Excel 文件、分析数据、修改现有表格、使用公式、生成图表时触发此技能。

帮我生成配图和插图

Sync files between Clawdbot workspace and Obsidian. Run the sync server to enable two-way file synchronization with the OpenClaw Obsidian plugin.

Comprehensive guide for creating professional UI/UX designs in Penpot using MCP tools. Use this skill when: (1) Creating new UI/UX designs for web, mobile, or desktop applications, (2) Building design systems with components and tokens, (3) Designing dashboards, forms, navigation, or landing pages, (4) Applying accessibility standards and best practices, (5) Following platform guidelines (iOS, Android, Material Design), (6) Reviewing or improving existing Penpot designs for usability. Triggers: "design a UI", "create interface", "build layout", "design dashboard", "create form", "design landing page", "make it accessible", "design system", "component library".

Interface for MCP (Model Context Protocol) servers via CLI. Use when you need to interact with external tools, APIs, or data sources through MCP servers, list available MCP servers/tools, or call MCP tools from command line.

Power BI semantic modeling assistant for building optimized data models. Use when working with Power BI semantic models, creating measures, designing star schemas, configuring relationships, implementing RLS, or optimizing model performance. Triggers on queries about DAX calculations, table relationships, dimension/fact table design, naming conventions, model documentation, cardinality, cross-filter direction, calculation groups, and data model best practices. Always connects to the active model first using power-bi-modeling MCP tools to understand the data structure before providing guidance.

Generate ASCII art diagrams using PlantUML text mode. Use when user asks to create ASCII diagrams, text-based diagrams, terminal-friendly diagrams, or mentions plantuml ascii, text diagram, ascii art diagram. Supports: Converting PlantUML diagrams to ASCII art, Creating sequence diagrams, class diagrams, flowcharts in ASCII format, Generating Unicode-enhanced ASCII art with -utxt flag

Oura Ring data integration and analytics. Fetch sleep scores, readiness, activity, HRV, and trends from the Oura Cloud API. Generate automated reports, correlations with productivity, and trigger-based alerts for low recovery days. Requires OURA_API_TOKEN (get at cloud.ouraring.com).

Interactive CLI testing specialist using tmux (Sonnet)

Deep dive into distributed tracing, APM tools, OpenTelemetry, logs aggregation, and observability best practices.

Comprehensive Stata reference for writing correct .do files, data management, econometrics, causal inference, graphics, Mata programming, and 17+ community packages (reghdfe, estout, did, rdrobust, etc.). Covers syntax, options, gotchas, and idiomatic patterns. Use this skill whenever the user asks you to write, debug, or explain Stata code.

当需要通过 HTTP 的 MCP 服务操作 GitHub（如创建/更新文件、批量提交、搜索仓库/代码/问题/用户、创建/查询/更新 Issue 与 PR、查看提交、审查与合并 PR 等）时使用。

This skill should be used when the user asks to "byted-codebase", "字节仓库", "codebase MCP", "查询代码库", or needs ByteDance codebase tooling.

使用 lark-cli add-permission 为飞书云文档添加协作者权限；适用于为用户/群/部门/群组/知识空间成员授予查看、编辑或管理权限。

通过检查最新 tag 和最近的 git 日志来创建 git 发布 tag，提出版本总结和 tag 消息，然后选择性地推送到远程。（Create git release tags by inspecting the latest tag and recent git log, proposing a version bump and tag message, then optionally pushing to remote.）

Repository structure and dependency analysis for understanding a codebase's architecture. Use when needing to (1) generate a file tree or structure map, (2) analyze import/dependency graphs, (3) identify entry points and module boundaries, (4) understand the overall layout of an unfamiliar codebase, or (5) prepare for deeper architectural analysis.

Interact with Safari browser tabs, reading list, bookmarks, and history via AppleScript. Use when the user asks to analyze, organize, summarize, deduplicate, close, export, or manage their Safari tabs. Also handles reading list, bookmarks, and history searches. Triggers include "my tabs", "open tabs", "Safari tabs", "clean up my browser", "what tabs do I have open", "organize my tabs", "too many tabs", "reading list", "bookmarks", "browser history", "export tabs". Requires macOS with Safari.

Convert Markdown documents (*.md files) to self-contained HTML files with embedded images. Use when you need a portable, offline-friendly single HTML file from Markdown—ideal for blog posts, essays, reports, or any content that should work without external dependencies.

Archive a Copilot CLI session to the brain. Creates a daily project file with the session transcript and adds a resume link to the weekly note. Use when ending a significant work session to preserve context for future reference.

Discovers home network devices via nmap, SSH, and DNS. Creates machine-readable inventory with topology diagrams. USE WHEN user says "scan home network", "discover home network", "map home network", "home network inventory", "homenet scan", "homenet discovery", "network topology", or scanning/mapping home network infrastructure.

Comprehensive document creation, editing, and analysis with support for tracked changes, comments, formatting preservation, and text extraction. When Claude needs to work with professional documents (.docx files) for: (1) Creating new documents, (2) Modifying or editing content, (3) Working with tracked changes, (4) Adding comments, or any other document tasks

Router for 3D building game mechanics in Three.js. Use when creating survival games, sandbox builders, base builders, or any game with player-constructed structures. Routes to 9 specialized skills for performance, physics, multiplayer, terrain, decay, UX, platform support, and design reference. Start here for building mechanics projects.

Presentation creation, editing, and analysis. When Claude needs to work with presentations (.pptx files) for: (1) Creating new presentations, (2) Modifying or editing content, (3) Working with layouts, (4) Adding comments or speaker notes, or any other presentation tasks

Format and lint GDScript files using gdscript-formatter. Use after editing GDScript files to ensure code style consistency.

| Reviewing | Read First | |-----------|------------| | Backend code | `.claude/shared/backend-patterns.md` | | Vue frontend | `.claude/shared/vue-frontend.md` | | React frontend | `.claude/shared/react-frontend.md` | | Tests | `.claude/shared/testing-patterns.md` | | Any code | `.claude/shared/critical-rules.md` |

Assists with memory analysis and forensics using MemProcFS. Use when analyzing memory dumps, investigating system activity, or performing forensic examinations.

规划和优化网站SEO关键词布局，检测关键词冲突，审查页面优化质量。Use when planning keyword strategy, diagnosing ranking issues, analyzing SERP competition, or when the user mentions "keyword conflicts", "internal competition", "keyword cannibalization", "page ranking problems", "keyword density", "SEO audit", or asks "why my pages don't rank", "how to distribute keywords", "which keywords for this page", "keyword layout strategy". 核心原则:一个页面一个主关键词+2-5个次要关键词。提供CSV检测脚本、布局模板和完整工作流程。

リリースワークフロー全体のガイド。CHANGELOG 形式、セマンティックバージョニング、Git フローのベストプラクティス。「release」「リリース」「version」「バージョン」「CHANGELOG」「tag」「GitHub Release」などのキーワードで自動適用。

Generate a production-ready SKILL.md + tiny examples, evals, and an index entry aligned to Anthropic Skills.

Search the web using Kagi's premium ad-free search engine, summarize web pages/PDFs/videos, and get AI-powered answers. Use when the user needs current information, wants to summarize a URL, or asks questions requiring web research. Kagi provides high-quality, unbiased results without SEO spam.

Supabase edge function sci_search for hybrid search over scientific journal chunks with optional journal/date filters, chunk expansion (extK), and metadata retrieval. Use when integrating or debugging sci_search requests, filters, or result shaping.

Use this skill when dispatching atomic, isolated development tasks to Jules, a remote AI coding agent.

Creates comprehensive handoff documents for seamless AI agent session transfers. Triggered when: (1) user requests handoff/memory/context save, (2) context window approaches capacity, (3) major task milestone completed, (4) work session ending, (5) user says 'save state', 'create handoff', 'I need to pause', 'context is getting full', (6) resuming work with 'load handoff', 'resume from', 'continue where we left off'. Proactively suggests handoffs after substantial work (multiple file edits, complex debugging, architecture decisions). Solves long-running agent context exhaustion by enabling fresh agents to continue with zero ambiguity.

Create factual working journal entries in Notes/WorkingJournal/ after completing analysis work. Use when user asks to "summarize work", "document results", or "create working journal entry". Ensures code is committed, copies figures to attachments, and creates objective summaries with citations.

Remote control Unity Editor via CLI using unityctl. Activate when user mentions Unity Editor, play mode, asset compilation, Unity console logs, C# script debugging, Unity tests, scene loading, or screenshots. Use for launching/stopping editor, entering/exiting play mode, compiling scripts, viewing logs, loading scenes, running tests, capturing screenshots, or executing arbitrary C# in Unity context.

Manage community member profiles to build persistent understanding of who community members are over time.

Automatically extract member profiles from synced Discord/Telegram messages using heuristic classification.

Standardize Git commit message format using Conventional Commits with Chinese subjects. Use when creating Git commits, checking commit message format, or preparing PRs. Strictly prohibit any AI identifiers in commit messages.

fraktor-rsのモジュール・型設計を「Less is more」「YAGNI」の観点でレビューし、過剰設計を検出する。参照実装（pekko/protoactor-go）との比較、定量分析、公開範囲の最適化を行う。トリガー：「型の設計をレビューして」「過剰設計チェック」「YAGNIチェック」「ファイル統合」「公開範囲を見直して」「type review」「overengineering check」等の型設計レビューリクエスト時に使用。

Idea → storyboard → Remotion overlays/animation → short renders + stills for fast visual iteration.

Create new Claude Code skills with proper structure, YAML frontmatter, and best practices. Use when creating reusable knowledge modules, adding specialized guidance, or building domain-specific expertise.

Execute and manage Rust tests including unit tests, integration tests, and doc tests. Use when running tests to ensure code quality and correctness.

STRICT GitHub release gatekeeper. Blocks premature releases (from develop, incomplete CI). Verifies PR merged to main + ALL CI passed before allowing tag/release. Triggers on "release", "tag", "publish", "deploy", "version".

Use when user asks to 'lint agent configs', 'validate skills', 'check CLAUDE.md', 'validate hooks', 'lint MCP'. Validates agent configuration files against 145 rules.

Shop on Rohlik.cz via MCPorter. Use to search products, manage cart, and place orders with explicit confirmation.

Knowledge management with QMD (Quick Markdown Search). Search, retrieve, and organize your personal knowledge base, notes, docs, and meeting transcripts. Use when asking about past notes, documentation, meetings, or when needing to recall information from your indexed markdown files.

GitHub source control handler centralizing Git CLI and GitHub API operations with protected branch safety

Create and manage Git worktrees for parallel development workflows

Preview infrastructure changes - run Terraform plan to show what resources will be created, modified, or destroyed. Generate human-readable plan summaries showing resource changes before deployment.

Shared utilities for faber-cloud plugin - configuration loading, pattern resolution, auto-discovery

Searches logs by content keywords, patterns, and filters with context extraction

Validates logs against type-specific schemas checking frontmatter, structure, and required fields

Creates or updates logs using type-specific templates with automatic validation and formatting

Manage milestones for release planning via Fractary CLI

[DEPRECATED] GitHub Issues handler - Use Fractary CLI instead

Post comments to issues via Fractary CLI with optional FABER context tracking

[DEPRECATED] Linear handler - Use Fractary CLI instead

[DEPRECATED] Jira Cloud handler - Use Fractary CLI instead

Shared utility skill for invoking fractary CLI codex commands

Creates or updates documentation in a type-agnostic manner by dynamically loading type-specific context at runtime

Handles multi-document operations with pattern matching and parallel execution, delegating to docs-manager-skill for each matched document

AWS S3 storage handler for fractary-file plugin

Prisma-specific database operations handler

Sync all projects in an organization with codex repository (with parallel execution). Delegates to fractary CLI for organization-wide sync operations.

Migrates codex configuration files from v2.0 push-based sync to v3.0 pull-based retrieval format with automatic backups

Performs comprehensive diagnostics on the codex cache system, detects issues, and can fix them automatically with repair operations. Delegates to fractary CLI for health checks and diagnostics.

HTTP/HTTPS handler for external documentation fetching. Fetches content from web URLs with safety checks and metadata extraction.

Sync a single project bidirectionally with codex core repository. Delegates to fractary CLI for sync operations.

Clear cache entries based on filters (all, expired, pattern). Delegates to fractary CLI for safe cache operations with dry-run support.

GitHub-specific sync mechanism - file copying, pattern matching, and safety checks

Analyzes the codex cache and generates comprehensive statistics about cache performance, storage usage, and health status. Delegates to fractary CLI for cache analysis operations.

Discover repositories in an organization for sync operations

Installs and configures custom status line functionality in Claude Code projects with proper plugin root variable usage

Validates skill responses against the standard FABER response format schema

Execute FABER phase hooks (pre/post phase boundaries)

<CONTEXT> This is a **prototype** for a deterministic workflow executor that addresses the "hallucinated completion" problem identified in Issue #346 investigation.

Load, validate, and resolve FABER configuration and workflows with inheritance support

Generates artifacts from templates by substituting {{VARIABLE}} placeholders with actual values. Uses template-engine.sh for deterministic variable substitution.

Detects hybrid agent anti-pattern where agents perform execution work directly instead of delegating to skills

Validates generated artifacts against Fractary plugin standards using automated validators. Checks XML markup, frontmatter, naming conventions, and structure.

Validates generated workflow components for compliance with Manager-as-Agent principle and correct architectural patterns

Outlines the Turborepo development workflow and patterns. Use as a reference for pnpm scripts and monorepo development tasks.

Implement tasks from an OpenSpec change. Use when the user wants to start implementing, continue implementation, or work through tasks.

Intercept deterministic tasks (math, time, currency, files, scheduling) BEFORE they hit the LLM. Saves 50-70% on token costs by resolving simple queries locally with zero API calls.

Pine Script v5/v6 indicator scaffold and patterns. Provides structure guidance and triggers doc-researcher for current syntax verification. Use when developing TradingView indicators.

Search PubMed database for biomedical research papers and retrieve titles, abstracts, and journal information

This skill should be used when users need to scrape websites, extract structured data, handle JavaScript-heavy pages, crawl multiple URLs, or build automated web data pipelines. Includes optimized extraction patterns with schema generation for efficient, LLM-free extraction.

Worker management for parallel task execution. Provides the `launch_worker` command-line tool for spawning worker agents in git worktrees. IMPORTANT: Always run launch_worker with `run_in_background=true` using the Bash tool so workers execute in parallel without blocking, the timeout should be 60 mins for safety. WORKER LIFECYCLE: 1. Progressive commit - worker commits all progress to the task branch 2. Completion - worker merges changes to Flow branch EXAMPLE - Launch single worker (~10 minute task): ``` # Create task branch for self-contained feature python -m flow_claude.scripts.create_task_branch \ --task-id="001" \ --instruction="Build user registration endpoint with validation. Use Write tool to create: 1) src/api/register.py with POST /api/register endpoint (accept email/password, validate email format, check password strength, hash password with bcrypt), 2) tests/test_register.py with tests for valid registration, duplicate email, weak password, invalid email format. Use Read tool to check existing src/models/user.py and src/utils/validation.py. Return proper JSON responses with 201 for success, 400 for validation errors. Estimated time: 10 minutes." \ --plan-branch="plan/add-user-features" \ --depends-on='[]' \ --context-paths='[]' # Create worktree git worktree add .worktrees/worker-1 task/001-build-registration-endpoint # Launch worker in background (use Bash tool with run_in_background=true) Bash( command="python -m flow_claude.scripts.launch_worker --worker-id=1 --task-branch='task/001-build-registration-endpoint' --cwd='.worktrees/worker-1' --plan-branch='plan/add-user-features' --model='sonnet'", run_in_background=true ) ``` EXAMPLE - Launch 3 parallel workers (each ~10 min, self-contained, independent): ``` # Task 001: Self-contained login endpoint (no dependencies on other tasks) python -m flow_claude.scripts.create_task_branch --task-id="001" --instruction="Create login endpoint. Use Write tool to create src/api/login.py with POST /api/login endpoint that accepts email/password, validates credentials against database, generates JWT token on success, returns token with 200 or error with 401. Use Read tool to check existing src/models/user.py for User model and src/utils/jwt.py for token generation. Write tests/test_login.py with tests for valid login, invalid password, non-existent user. ~10 minutes." --plan-branch="plan/add-user-features" --depends-on='[]' --context-paths='[]' # Task 002: Self-contained profile endpoint (independent from login/register) python -m flow_claude.scripts.create_task_branch --task-id="002" --instruction="Create user profile endpoints. Use Write tool to create src/api/profile.py with GET /api/profile/:id to fetch user data and PUT /api/profile/:id to update user profile (name, bio, avatar_url). Add authentication check using existing JWT middleware. Use Read tool to check src/models/user.py. Write tests/test_profile.py with tests for get profile, update profile, unauthorized access. ~10 minutes." --plan-branch="plan/add-user-features" --depends-on='[]' --context-paths='[]' # Task 003: Self-contained password reset (independent from other auth features) python -m flow_claude.scripts.create_task_branch --task-id="003" --instruction="Create password reset flow. Use Write tool to create src/api/password_reset.py with POST /api/password-reset/request (send reset email with token) and POST /api/password-reset/confirm (verify token and update password). Use Read tool to check existing src/services/email.py for sending emails and src/models/user.py. Write tests/test_password_reset.py with tests for request reset, invalid email, confirm reset, expired token. ~10 minutes." --plan-branch="plan/add-user-features" --depends-on='[]' --context-paths='[]' # Create worktrees git worktree add .worktrees/worker-1 task/001-create-login-endpoint git worktree add .worktrees/worker-2 task/002-create-profile-endpoints git worktree add .worktrees/worker-3 task/003-create-password-reset-flow # Launch all workers in parallel - each task is self-contained and independent Bash(command="python -m flow_claude.scripts.launch_worker --worker-id=1 --task-branch='task/001-create-login-endpoint' --cwd='.worktrees/worker-1' --plan-branch='plan/add-user-features' --model='sonnet'", run_in_background=true) Bash(command="python -m flow_claude.scripts.launch_worker --worker-id=2 --task-branch='task/002-create-profile-endpoints' --cwd='.worktrees/worker-2' --plan-branch='plan/add-user-features' --model='sonnet'", run_in_background=true) Bash(command="python -m flow_claude.scripts.launch_worker --worker-id=3 --task-branch='task/003-create-password-reset-flow' --cwd='.worktrees/worker-3' --plan-branch='plan/add-user-features' --model='sonnet'", run_in_background=true) ```

Symfony 7.4 Asset component reference for managing web asset URLs and versioning. Use when working with asset versioning, CDN configuration, asset packages, static file URL generation, or cache busting. Triggers on: Asset component, asset versioning, CDN, asset packages, JsonManifestVersionStrategy, static assets, UrlPackage, PathPackage, VersionStrategy, Packages class.

Symfony 7.4 Clock component reference for decoupling applications from the system clock and testing time-sensitive logic. Use when working with Clock, ClockInterface, MockClock, NativeClock, MonotonicClock, DatePoint, ClockAwareTrait, ClockSensitiveTrait, time testing, now() helper, or Doctrine date_point/day_point/time_point types. Triggers on: Clock, ClockInterface, MockClock, NativeClock, DatePoint, time testing, ClockAwareTrait, ClockSensitiveTrait, now().

Symfony 7.4 ExpressionLanguage component reference for compiling and evaluating expressions. Use when working with expression evaluation, expression compilation, custom expression functions, expression providers, expression caching, or any ExpressionLanguage-related Symfony code. Triggers on: ExpressionLanguage, expression evaluation, expression compilation, custom functions, ExpressionFunctionProviderInterface, ParsedExpression, expression syntax, evaluate(), compile(), expression linting.

Symfony 7.4 Mime component reference for creating and manipulating MIME messages. Use when working with email messages, MIME types, file attachments, embedded images, or email content. Triggers on: Mime, MIME types, email messages, Email class, multipart, attachments, MimeTypes, Address, TemplatedEmail, RawMessage, Message, Headers, embedded images, DataPart, TextPart, multipart/mixed, multipart/alternative.

Symfony 7.4 Finder component reference for finding files and directories with an intuitive fluent interface. Use when searching for files or directories, filtering by name, content, size, date, depth, path, glob patterns, or sorting results. Triggers on: Finder, file search, directory search, glob, name filtering, content filtering, sorting, date filtering, size filtering, depth filtering, path filtering, SplFileInfo, ignoreVCS, followLinks.

Symfony 7.4 UID component reference for generating and manipulating unique identifiers. Use when working with UID, UUID, ULID, unique identifiers, UuidV4, UuidV7, AbstractUid, Doctrine UUID type, UuidFactory, Ulid, toBase32, toBase58, toRfc4122, NilUuid, NilUlid, UlidFactory, UuidType, UlidType, or any unique identifier-related Symfony code.