AI Skills Hub - Browse 180,000+ Agent Skills

Train reinforcement learning agents using AgentLightning/VERL with PPO. Validates data, configures training, executes training runs. Use for agent training, RL training, PPO, reinforcement learning, VERL.

使用 AlphaEvolve/OpenEvolve/alphaevolve-run 执行MLE代码进化或 prompt 进化任务。支持 MLE 代码进化（如机器学习模型优化）和 Prompt 进化（如提示词优化）。触发场景：启动 alphaevolve 进化、运行 openevolve、MLE 代码进化、prompt 进化、使用 alphaevolve 运行任务。本 skill 提供完整的输入校验、文件共创、运行管理和结果检查功能。最终启动进化时按照步骤4的bash命令来运行（AlphaEvolve/OpenEvolve已经封装好）

Read, write, and update Google Sheets data via CLI. Use when the user asks to read spreadsheet data, update cells, append rows, or work with Google Sheets. Triggers on mentions of spreadsheets, sheets, Google Sheets, tabular data in the cloud, or specific sheet names like "Projects" or "Tasks".

Deterministic wrapper for tonl CLI (npm package) for structured data operations in JSON/TONL formats. Use when converting JSON↔TONL, querying structured data, validating TONL schemas, or generating statistics. Maps directly to CLI subcommands (encode, decode, query, get, validate, stats).

MUST BE USED when creating, improving, or reviewing Skills. Use PROACTIVELY when user mentions "create a skill", "new skill", "write a skill", "improve skill", "skill description", or "progressive disclosure". Guides through specification-compliant skill design with three-level progressive disclosure.

使用 pandoc 将 Markdown 文件转换为 Jira/Confluence wiki 格式。 适用场景：(1) 将 .md 文件转换为 wiki 格式 (2) markdown 转 jira 格式 (3) 生成 confluence 兼容的 wiki 文件。 用户需在提示词中指定要转换的 markdown 文件路径。

Analyze and validate forms on web pages. Use when users ask to check form accessibility, verify form labels, audit input fields, check form validation, or analyze form UX. Detects missing labels, invalid input types, accessibility issues, and validation problems.

Analyze and validate meta tags on web pages. Use when users ask to check meta tags, verify SEO tags, audit page titles, check Open Graph tags, verify canonical URLs, or analyze social sharing tags. Detects missing title, description issues, duplicate tags, and Open Graph problems.

Analyze performance-related HTML issues on web pages. Use when users ask to check page performance, audit resource loading, verify lazy loading, check preload hints, or analyze render-blocking resources. Detects missing optimizations, large resources, render-blocking issues, and Core Web Vitals problems.

Analyze and validate images on web pages. Use when users ask to check image alt tags, verify image accessibility, find missing alt attributes, audit image SEO, or check image optimization. Detects missing alt, empty alt, decorative images without proper markup, oversized images, and missing dimensions.

Analyze and validate links and buttons on web pages. Use when users ask to check links, verify button hrefs, find empty links, detect javascript:void links, find broken navigation, or audit clickable elements. Detects empty href, javascript:void(0), hash-only links, missing href attributes, and non-accessible buttons.

Perform accessibility audit on web pages. Use when users ask to check accessibility, audit WCAG compliance, verify ARIA usage, check color contrast, audit keyboard navigation, or analyze screen reader compatibility. Detects ARIA issues, focus problems, contrast issues, and semantic HTML violations.

Analyze and validate HTML heading hierarchy (H1-H6) on web pages. Use when users ask to check headings structure, verify H1 presence, analyze SEO heading hierarchy, find heading issues, or audit page structure. Detects missing H1, multiple H1s, skipped levels, empty headings, and hierarchy violations.

The pipeline uses a **two-phase approach** for effects with temporal envelopes:

spawn_client = node.create_client(SpawnEntity, '/spawn_entity') request = SpawnEntity.Request() request.name = 'my_humanoid' request.xml = open('humanoid.urdf', 'r').read() request.initial_pose.position.z = 1.0 spawn_client.call_async(request) ```

Oligon scientific brand implementation for figures, presentations, and documents. Provides color palettes, typography standards, and format-specific adapters for matplotlib, ggplot2, ReportLab, python-docx, python-pptx, and HTML/CSS.

Convert templated markdown to branded PDFs. Create documents from 12 templates (scientific, project-mgmt, technical), generate professional PDFs with Oligon styling.

Generate and edit images using AI models (FLUX, Gemini) via OpenRouter. For photos, illustrations, artwork, and visual assets. Use scientific-schematics for technical diagrams.

Comprehensive citation management for academic research: search databases, extract metadata, validate citations, generate formatted BibTeX entries.

Write a daily or weekly work report using git commits. Use when the user asks to write or send a daily report/standup or weekly report, especially "日报", "发日报", "周报", "发周报", "daily report", "weekly report", or "work report".

A strict protocol for using Context7 tools to fetch accurate, up-to-date documentation without hallucinating library IDs.

日本語テキストの誤字脱字チェック、表記揺れの検出と修正。小説、技術文書、ブログ記事などで使用。ユーザーが「誤字脱字をチェック」「表記の揺れを確認」「校正して」などと依頼した時、または文書の品質向上が必要な時に使用。

Automated Kafka cluster provisioning for Phase V event-driven architecture using Strimzi or Redpanda.

WiX MSI インストーラーで Windows Service のアップグレード時にハングする問題のトラブルシューティング

Orchestrates multi-model LLM consensus through a three-phase deliberation protocol. Use when you need collaborative AI review, multi-model problem-solving, code review from multiple perspectives, or consensus-based decision making. Coordinates OpenAI Codex, Google Gemini, and Claude CLIs for opinion collection, peer review, and chairman synthesis.

Automate release process - bump version, update CHANGELOG, push commits. Auto-tag workflow handles tagging. Use when user asks to create/publish a release, or says "/release"

Use when about to claim PM deliverable is complete, before finalizing documents, or before stakeholder review. Requires checking success criteria with evidence before any completion claims.

Use when building new skills, creating SKILL.md files, or setting up skill structure.

Keď používateľ pýta o fungovaní modelu alebo dôveryhodnosti predikcií, postupuj nasledovne:

태스크 파일 CRUD, 상태 업데이트, 아카이빙 가이드

복잡한 판단이 필요할 때 decision-agent 활용 가이드

Search for items and properties on Wikidata and retrieve entity details, claims, and external identifiers. Supports both keyword search (Wikidata Action API) and semantic/hybrid search (Wikidata Vector Database), plus direct entity retrieval (Special:EntityData) and structured querying (WDQS SPARQL).

Render the book into HTML or PDF formats using Quarto. Use when the user wants to "render the book", "build the book", "generate the book", or create HTML/PDF output.

Merge a cover image into a PDF book while preserving aspect ratio and matching width. Use when the user wants to "merge cover", "combine pdf", "fix cover size", or "add cover image".

Extract hyper-relevant findings from academic PDFs with killer precision. Focused on computational social science, LLM agents, online communities, network science, information theory, and statistical physics. Outputs brief, citable notes on novel methods, measures, benchmarks, and use cases.

Full implementation mode - end-to-end feature implementation with phased execution, parallel work streams, verification gates, and atomic commits per phase

Guide systematic debugging through hypothesis generation, evidence collection, and verification. Uses scientific method for root cause analysis.

Rapid research prototyping through mini-experiments. A meta-skill for exploring ideas quickly with time-boxed experiments, toy datasets, and structured decision-making.

Show detailed information about a bead. Use when checking bead details, when the user asks "show bead X", "what is bead X", "bd show X".

Create a new bead. Use when creating tasks, when the user asks "create a task", "add a bead", "bd create".

Conversational methodology for creating comprehensive product definition documents. Use when the user wants to create a product spec, PRD, or product definition file for a new project or feature.

Find duplicate type names within your solution (same name in different namespaces). Use before creating new types to avoid naming conflicts. Only checks your code, not external packages. (project, gitignored)

Run `dotnet publish` with Native AOT enabled and analyze trim/linker warnings (IL2xxx/IL3xxx).

Analyze C# code complexity to identify methods that need testing. Use when you need to understand code complexity or prioritize testing efforts. (project, gitignored)

Validates that all `JsonSerializer` usage is AOT-compatible by checking for proper `JsonSerializerContext` usage.

List all types within a specific namespace or explore namespace structure. Use when you need to understand what's in a namespace or plan where to add new types. (project, gitignored)

Manage content filtering rules for the DNR newsletter. Add or update exclusion keywords, test if headlines would be filtered, identify false positives/negatives, and understand current filter behavior.

Diagnose RSS feed health issues, identify broken or underperforming feeds, and suggest fixes or alternatives. Use when feeds return no stories, when major outlets seem missing from newsletters, or for periodic feed audits.

Manage news source configuration for the DNR newsletter. Add new outlets, update domain mappings for source attribution, validate source names, and handle source-related issues like missing attribution.

Comprehensive Python code review and validation toolkit for assessing code quality, identifying issues, and ensuring best practices. When Claude needs to: (1) Review Python code for PEP8 compliance and style issues, (2) Identify potential bugs, security vulnerabilities, and edge cases, (3) Check for proper error handling and documentation, (4) Validate test coverage and code structure, or (5) Suggest performance and architectural improvements. Use for code review and quality assurance, not for writing new code from scratch.

深度代码审查工具：分析 Git 变更，评估逻辑、代码规范、安全及需求一致性，支持CodeReview报告。

Access shared business context including company identity, mission, org chart, staff directory, and policies. Use this when an agent needs to understand the business they work for, organizational structure, who's who, or business policies. Returns content from org/business/ directory.

Process multimedia files with FFmpeg (video/audio encoding, conversion, streaming, filtering, hardware acceleration), ImageMagick (image manipulation, format conversion, batch processing, effects, composition), and RMBG (AI-powered background removal). Use when converting media formats, encoding videos with specific codecs (H.264, H.265, VP9), resizing/cropping images, removing backgrounds from images, extracting audio from video, applying filters and effects, optimizing file sizes, creating streaming manifests (HLS/DASH), generating thumbnails, batch processing images, creating composite images, or implementing media processing pipelines. Supports 100+ formats, hardware acceleration (NVENC, QSV), and complex filtergraphs.

Search technical documentation using executable scripts to detect query type, fetch from llms.txt sources (context7.com), and analyze results. Use when user needs: (1) Topic-specific documentation (features/components/concepts), (2) Library/framework documentation, (3) GitHub repository analysis, (4) Documentation discovery with automated agent distribution strategy

OBS Optimizer プロジェクトのRustバックエンド開発用スキル。 このスキルは、Rustファイル(src-tauri/src/)の編集時にTDD駆動でベストプラクティスに従った実装とセルフコードレビューを自動的に行う。 バックエンドの機能実装・改善、新規Tauriコマンド追加時に使用する。

Guide interactions with the human Super User/Operator who performs external actions, approvals, and evidence collection. Use when Codex needs privileged access actions, environment confirmations, command output, screenshots, or approval for risky operations.

Expert guide for using n8n-mcp MCP tools effectively. Use when searching for nodes, validating configurations, accessing templates, managing workflows, or using any n8n-mcp tool. Provides tool selection guidance, parameter formats, and common patterns.

CodeRabbit AI 리뷰 코멘트를 자동으로 처리하는 스킬입니다.

Search and retrieve information from indexed PDF documentation including IC datasheets, FPGA manuals, and technical specifications. Use this when the user asks about hardware specifications, pin configurations, register details, timing diagrams, or any technical information that might be in datasheets or technical documentation.

Creation and maintenance of financial import strategies (PDF, OFX, CSV, etc.) using the Strategy pattern.

Pick up User Story work items from Azure DevOps, assign to developer, create git branch, investigate, plan, and implement the feature

Task 사용자 매뉴얼(080-user-manual.md) 생성 스킬. Playwright MCP로 스크린샷 캡쳐, 템플릿 기반 문서 작성을 자동화합니다. 트리거: "사용자 매뉴얼 생성", "080-user-manual 작성", "스크린샷 캡쳐해서 매뉴얼", "Task 사용자 가이드 작성", "user-manual-creator 실행"

Initialize jjiban project management directory structure. Use when creating .jjiban folder with settings, templates, and project folders. Triggers on "init jjiban", "create jjiban structure", "setup project management", "initialize .jjiban", or "new jjiban project".

Pick up Bug work items from Azure DevOps, assign to developer, create git branch, investigate, plan, and implement the fix

File-Based Apps는 .NET 10에서 도입된 기능으로, 단일 C# 파일에서 프로젝트 파일(.csproj) 없이 .NET 애플리케이션을 빌드, 실행, 게시할 수 있게 해준다.

Serve Jekyll site accessible via VPN tunnel and Playwright. Use when the user asks to browse the Jekyll site, visit council-test.cyberknight-websites.com URLs, or access the Jekyll server with Playwright.

Reference for shared utility modules. Use when working with Git commands, GitHub operations, or need API details for utility functions.

Configures Azure Static Web Apps routing, authentication, deployment, and troubleshoots SWA-specific issues. Use for SWA-related problems, auth setup, route configuration, and production deployment.

description: | プ譜（プロジェクト譜）のJSONデータを既存のドキュメントから自動生成するスキル。 使用するタイミング： (1) PowerPoint、Excel、PDF、テキストファイルからプ譜を生成したい (2) 複数のドキュメントを解析してプ譜の要素を抽出したい (3) プロジェクト関連資料からプ譜を自動作成したい (4) プ譜エディタ（pufu-editor）で使えるJSON形式でエクスポートしたい

Create and run database migrations with Alembic

Create a new Claude Code skill for automating repetitive tasks

Access Google Workspace APIs (Gmail, Drive, Sheets, Docs) via oauth2l + curl.

Use this skill when the user asks to review the security of plugins, scan plugins for vulnerabilities, or audit installed plugins. Also suggest this skill when you notice a new plugin has been added.

Create custom tools for the vibe agent. Use when user wants to "add a tool", "create a tool", "extend vibe capabilities", or mentions "BaseTool", "ToolArgs", "ToolResult".

MCP (Model Context Protocol) integration patterns. Use when working with MCP servers, adding external tool integrations, or when user mentions "MCP", "stdio transport", "http transport", "external tools", or "tool proxy".

**READ FIRST** before ANY Dataverse/Power Platform MCP tool calls (get-record, query-records, create-record, update-record, delete-record). Covers field types, lookup syntax, and common gotchas. Triggers: create, clone, update, delete, link records, CRM URLs (crm11.dynamics.com), or any entity operations (contacts, accounts, si_*, sic_* tables).

Create and update Azure DevOps Wiki pages with consistent formatting. Use when asked to create wiki page, write wiki documentation, update wiki article, or work with ADO wiki. Includes page structure templates, formatting syntax, and API versioning requirements.

This skill should be used when the user asks to "create a CLAUDE.md", "write a CLAUDE.md", "set up CLAUDE.md", "configure Claude for this project", "add project instructions for Claude", "initialize Claude context", or mentions needing project-specific Claude instructions.

Robust Rust patterns for file-backed data, parsing, persistence, FFI boundaries, and system integration. Use when writing Rust that handles file formats, subprocess integration, PID/process management, Serde serialization, or UniFFI boundaries. Covers UTF-8 safety, atomic writes, state machines, and defensive error handling.

Expertise in architecting, implementing, reviewing, and debugging hierarchical matching systems. Use when working with: (1) Two-sided matching (Gale-Shapley, hospital-resident, student-school), (2) Assignment/optimization problems (Hungarian algorithm, bipartite matching), (3) Multi-level hierarchy matching (org charts, taxonomies, nested categories), (4) Entity resolution and record linkage across hierarchies. Triggers: debugging match quality issues, reviewing matching algorithms, translating business requirements into constraints, validating match correctness, architecting new matching systems, fixing unstable matches, resolving constraint violations, diagnosing preference misalignment.

Guide users step-by-step through manually testing whatever is currently being worked on. Use when asked to "test this", "verify it works", "let's test", "manual testing", "QA this", "check if it works", or after implementing a feature that needs verification before proceeding.

一次性配置SkillForge的marketplace设置

지피터스(gpters.org) 멤버 프로필에서 게시물을 스크래핑하여 Markdown으로 저장. "지피터스 게시물 수집", "gpters 스크랩", "editor_소연 게시물", "프로필 크롤링" 등을 언급하면 자동 실행. Selenium 기반 동적 페이지 수집.

Guiar el desarrollo, debugging y mantenimiento del pipeline DB-centric de procesamiento de fondos.

Systematic debugging of code issues and errors. Use when encountering errors, test failures, runtime exceptions, type errors, or unexpected behavior. Triggers on error messages, stack traces, or when something isn't working as expected.

Weekly Focus Engineering analysis using ActivityWatch data. Use when analyzing app usage patterns, detecting context switching problems, identifying "death loops" (repetitive app switching), calculating focus scores, or creating weekly productivity reviews.

Help address review/issue comments on the open GitHub PR for the current branch using gh CLI; verify gh auth first and prompt the user to authenticate if not logged in.

Install Codex skills into $CODEX_HOME/skills from a curated list or a GitHub repo path. Use when a user asks to list installable skills, install a curated skill, or install a skill from another repo (including private repos).

No description available

No description available

创建或修改 CRUD 模块。当需要创建新的增删改查 API、修改现有路由模块、添加新字段、新增接口、或用户请求"创建/修改 XX 管理"时使用

创建或修改数据库 Schema。当需要创建新表、修改表结构、定义字段、设置索引约束、或涉及 Drizzle ORM / drizzle-zod 操作时使用

Testing strategies for the Igloo Insurance platform including Property-Based Testing with Hypothesis. Use when writing tests for calculation engines, verifying financial invariants (payout never exceeds 100%), testing concurrent control mechanisms (Redis locks, DB locks), or validating timezone-sensitive logic. Covers pytest-asyncio patterns and edge case discovery.

CPU-bound calculation engine patterns for Risk and Claim calculations in the Igloo Insurance platform. Use when implementing risk event detection, tier differential claim calculations, timezone conversions for "per day/month" logic, or concurrent control with Redis locks. Covers financial precision with Decimal, async execution strategies, and distributed locking.

Comprehensive spreadsheet creation, editing, and analysis with support for formulas, formatting, data analysis, and visualization. When Claude needs to work with spreadsheets (.xlsx, .xlsm, .csv, .tsv, etc) for: (1) Creating new spreadsheets with formulas and formatting, (2) Reading or analyzing data, (3) Modify existing spreadsheets while preserving formulas, (4) Data analysis and visualization in spreadsheets, or (5) Recalculating formulas

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

Create, read, and edit PowerPoint presentations (.pptx files). Use when building slide decks, extracting content from presentations, or automating presentation creation.

Read, write, and manipulate Excel spreadsheets (.xlsx files). Use when working with spreadsheet data, creating reports, extracting data from Excel files, or automating spreadsheet tasks.

Create, read, and edit Word documents (.docx files). Use when working with professional documents, reports, creating formatted text documents, or extracting text from Word files.

Backend開発時に必要な検証を実施するスキルです。Lint、ビルド、ユニットテスト、統合テスト、OpenAPI仕様生成を含みます。Backend（apps/backend）のコード変更時に使用してください。