AI Skills Hub - Browse 180,000+ Agent Skills

Extract text from images using pytesseract OCR. Handles PNG, JPG, JPEG, GIF, BMP, TIFF formats. Use when processing image files that may contain text, diagrams with labels, screenshots, or scanned documents. Returns extracted text content. Routed from file-identifier.

Extract PostgreSQL database content to text-readable formats (CSV, JSON, Markdown) for LLM processing and content synthesis. Use when working with PostgreSQL databases that need to be converted to text formats for analysis, synthesis, or inclusion in LLM context windows. Routed from db-router skill.

Identify database file formats by scanning magic bytes and file signatures. Use FIRST when encountering any unknown database file. Returns format type, version, confidence level, and DBeaver CLI compatibility. Pure identification only - for routing decisions, pass results to db-router.

Route database files to appropriate extraction method based on identification from db-identify. Decides between DBeaver CLI (for supported formats) or db-extractor-* skills (for unsupported formats). Use AFTER db-identify to determine extraction workflow.

Extract MySQL database content to text-readable formats (CSV, JSON, Markdown) for LLM processing and content synthesis. Use when working with MySQL databases that need to be converted to text formats for analysis, synthesis, or inclusion in LLM context windows. Routed from db-router skill.

Extract SQLite database content to text-readable formats (CSV, JSON, Markdown) for LLM processing and content synthesis. Use when working with SQLite databases (.db, .sqlite, .sqlite3) from course materials, datasets, or applications that need to be converted to text formats for analysis, synthesis, or inclusion in LLM context windows. Routed from db-router skill.

Unified file identification and routing for KnowledgeSynthesizer pipeline. Detects file types via magic bytes with extension fallback, returns extractor skill to invoke. Use FIRST when processing any unknown file. Handles databases (sqlite, access), PDFs, images (png, jpg, gif, bmp, tiff), HTML, archives (zip, rar, 7z, tar), Office docs (docx, pptx, xlsx), and plain text.

Extract archive files using 7z CLI. Handles ZIP, RAR, 7z, TAR, TAR.GZ, TAR.BZ2 formats. Extracts to subfolder named after archive. Returns list of extracted files for further processing by file-router. Use when processing compressed archives containing course materials or mixed file types.

This skill should be used when the user asks to "generate an image", "create an image", "edit an image", "transform this image", "apply style to an image", "make a picture of", "create variations", or mentions "Gemini image", "Nano Banana", or any image generation/editing tasks. Provides guidance for using Google's Gemini image models.

Start and manage OpenSearch server with Docker. Use when starting OpenSearch locally, setting up development environment, or managing OpenSearch containers.

AI-powered IMAP email client with semantic search, intelligent triage, and graduated autonomy. Manages email through natural conversation.

Guides through BOSS numbered development workflow (feature branch → quality check → changeset → PR). Use when starting features, running checks, creating changesets, or submitting PRs.

Execute GitHub CLI operations with 1Password credential management. Use when working with PRs, workflows, issues, or any gh command requiring authentication.

Use this skill when integrating an external Go project with a production consent identity server. Triggers include setting up authentication, protecting routes with tokens, handling OAuth authorization codes, configuring ECDSA public keys, implementing CSRF protection, or connecting to a consent server.

Playwright-BDD/Cucumberテストの品質分析スキル。 以下の場合に使用: (1) .featureファイルのレビュー・品質チェック時 (2) ステップ定義の再利用性分析時 (3) シナリオ網羅性の確認時 (4) E2Eテスト実行結果の分析時 (5) テストの保守性評価時

Executes implementation tasks from a plan. Use when executing a single task from an implementation plan with TDD workflow.

阿里云 ECS 云服务器管理 - 实例查询、状态监控

部署状态监控 - 实时跟踪部署进度、检测异常、输出日志

AI 图片生成 - 支持 Gemini、即梦、通义万相、Grok 多 Provider

AI 视频生成 - 支持 Veo 3.1、即梦、通义万相、Grok 多 Provider

Terraform infrastructure code review, validation, and best practices. Activates when reviewing .tf files, analyzing terraform plan output, discussing IaC patterns, or troubleshooting Terraform errors. Covers modules, state management, security, and cost optimization.

ALWAYS USE FIRST when exploring, searching, or researching this codebase. Use symbol_tree MCP tool before grep/glob/read. Provides instant architecture overview, function relationships, and call graphs without reading full implementations.

Use this skill when the user asks to create a new skill, custom command, or wants to know how skills work in Claude Code.

Kubernetes troubleshooting, diagnostics, and incident response. Activates when debugging pod failures, analyzing cluster issues, reviewing K8s manifests, or responding to production incidents. Covers deployments, services, networking, and resource management.

Build desktop apps with GPUI, the GPU-accelerated UI framework from Zed. Covers Entity state, Render trait, div() Tailwind API, actions/keybindings, gpui-component widgets, theming. Use when building Rust desktop applications with GPUI or gpui-component.

Window management APIs for Script Kit GPUI using macOS Accessibility (AX) APIs. Use when working with: window positioning/tiling, window state persistence, display/monitor detection, window capabilities, resize coalescing, or Space (virtual desktop) operations. Covers modules: window_control.rs, window_control_enhanced/, window_manager.rs, window_ops.rs, window_resize.rs, window_state.rs.

Unified icon system for Script Kit GPUI. Use when working with icons in the UI - rendering Lucide icons, embedded SVGs, SF Symbols (macOS), app bundle icons, or custom file/URL icons. Covers IconRef parsing, IconStyle configuration, color tokens, and the IconView rendering API.

GPUI framework patterns for Script Kit. Use when writing UI code, handling keyboard events, managing state, or working with layouts. Covers layout chains, lists, themes, events, focus, and window management.

Logging and observability patterns for Script Kit GPUI. Use when adding logs, debugging, or understanding the logging system. Covers JSONL format, compact AI log mode, correlation IDs, and error handling.

Script metadata and scriptlet bundle patterns for Script Kit. Use when writing scripts, test files, or scriptlet bundles. Covers global metadata, legacy comment-based metadata, and YAML frontmatter for bundles.

Script Kit GPUI codebase architecture. Use when navigating the codebase, understanding module structure, or tracing data flow. Covers repository structure, SDK deployment, configuration, and secondary windows.

Hive/beads task management for Script Kit agents. Use when working with the issue tracking system, managing beads, or coordinating agent work. Covers issue lifecycle, progress reporting, and file reservations.

Open files edited by the agent in VSCode after making changes, using the bundled open-changed-files script (silent no-op when unavailable).

Create a traceable progress planning file under docs/progress/ from PROGRESS_TEMPLATE.md, capturing goals/acceptance criteria/scope/I-O contract/decisions/risks/step checklist, then open a GitHub PR with gh (no feature implementation yet). Use when the user and agent have aligned on a new feature and want an executable plan to guide future work and review.

Merge and close a feature PR with gh after a quick PR hygiene review (title, required sections, testing notes) aligned with create-feature-pr. Use when the user asks to merge/close a feature PR, delete the remote branch, and do post-merge cleanup.

Finalize and archive a progress file for a GitHub PR: locate the related docs/progress file (prefer PR body Progress link), set Status to DONE, move it to docs/progress/archived/, update docs/progress/README.md, merge the PR with gh, patch the PR body Progress link to point to the base branch so it survives branch deletion, and (when present) patch the planning PR body to include an Implementation PR link. Use when a feature PR is ready to be closed and its progress tracking should be marked DONE.

This skill should be used when modifying the database schema, running Prisma migrations, querying the database with Prisma ORM, or working with PostgreSQL in the ClaudeCode Sentiment Monitor project. Specifically trigger this skill for tasks involving schema changes, migrations, Prisma client operations, or database structure questions.

Efficiently complete web research and content analysis tasks using search and read tools. Use when the user needs to research topics, find latest information, compare products, or extract content from web pages.

Build real-time voice and video applications with Google's Gemini Live API. Use when implementing bidirectional audio/video streaming, voice assistants, conversational AI with interruption handling, or any application requiring low-latency multimodal interaction with Gemini models. Covers WebSocket streaming, voice activity detection (VAD), function calling during conversations, session management/resumption, and ephemeral tokens for secure client-side connections.

Guidelines for creating and organizing scripts in this project. Use when creating new scripts, deciding where scripts should live, or cleaning up temporary work. Covers tmp/ for scratch work vs scripts/ for reusable tools.

Run code quality checks including duplication detection (jscpd), unused dependency analysis (depcheck), and complexity metrics. Use when the user asks to check code quality, find duplicate code, detect copy-paste, find unused packages, or analyze code complexity.

Write migrations in `src/lib/migrations/index.ts`.

Ubuntu server configuration via cloud-init and systemd. Use when writing cloud-init YAML, configuring systemd services, setting up nginx, or troubleshooting VM configuration issues.

Diagnose server-side issues using JSONL logs. Use when debugging slow operations, failed requests, migration problems, data corruption, or understanding what happened during a user session. Triggers: performance issues, 500 errors, 'why is this slow', plan loading problems, undo/redo bugs, checkpoint issues, migration failures.

Query TypeScript codebase structure. Use when asking "what does X contain?", "where is X defined?", "what calls X?", or exploring type/interface/function definitions. Answers structural questions by parsing actual code.

Build comprehensive, mobile-compatible Obsidian study vaults from academic course materials with checkpoint-based workflow, error pattern recognition, and quality assurance. Battle-tested patterns from 828KB/37-file projects. Works across all subjects - CS, medicine, business, self-study.

This skill MUST be invoked when the user says "evaluate alternatives", "make technology choice", "document decision", "technology choice", "trade-offs", "decision record", "rationale", or "why we chose". SHOULD also invoke when user mentions "alternatives" or "NEEDS CLARIFICATION".

This skill MUST be invoked when the user says "design API", "map endpoints", "define schemas", "API contract", "REST API design", or "OpenAPI spec". SHOULD also invoke when user mentions "endpoint", "schema", "contract", or "HTTP".

This skill MUST be invoked when the user says "write requirements", "define success criteria", "identify edge cases", or "functional requirements". SHOULD also invoke when user mentions "FR-", "SC-", "RFC 2119", "MUST SHOULD MAY", or "edge cases". Produces technology-agnostic requirements in FR-XXX format with measurable success criteria.

将小说章节/剧本拆解为 Sora 2 可用的分镜脚本。Use when user asks to create storyboard, breakdown scenes, 分镜, 拆解场景, 场景卡片, or convert script/novel to video shots.

Write and optimize SPARQL queries for RDF data extraction. Use when creating queries for ggen transformations, extracting data from TTL files, or optimizing query performance.

Validate RDF/Turtle syntax and SHACL shape constraints. Use when checking TTL files for syntax errors, validating against SHACL shapes, or ensuring ontology compliance.

Detects Python project configuration and provides commands for testing, building, coverage, and containerization. Use when: Starting workflow, detecting project stack, need TEST_CMD Triggers: detect stack, what commands, initial setup Outputs: TEST_CMD, BUILD_CMD, COVERAGE_CMD, COVERAGE_CHECK, MIGRATE_CMD

Manages git operations: worktree creation, branch management, commits, PRs, semantic versioning, daily rebase workflow, and PR feedback handling. Use when: Creating branches/worktrees, committing, pushing, versioning, handling PR feedback Triggers: create worktree, commit, push, rebase, version, PR, PR feedback

Railway deployment patterns for Campreserv. Use when deploying, debugging deployments, or managing Railway services.

Expert knowledge of proven product management frameworks for discovery, growth, measurement, planning, and AI-era practices.

Develop and design SDKs for web APIs (REST/GraphQL/OpenAPI), including API surface design, client architecture, auth, errors, retries, pagination, and release practices. Use this skill when asked to create, refactor, or review an SDK, generate language-specific clients, or produce SDK guidelines, checklists, and docs for API consumption.

Monitor X/Twitter for high-engagement trending posts matching specific keywords or from followed accounts. This skill detects posts with rapid engagement growth (e.g., 1000+ likes/hour) or high absolute engagement (e.g., 5000+ total interactions), then sends email notifications. Use this skill when setting up automated X/Twitter monitoring for marketing, content opportunities, or trend detection.

Knowledge base for MS365 operations using the @softeria/ms-365-mcp-server.

Git-backed issue tracking for AI agent workflows. Replaces handoff/catch-up with persistent execution state. Use for session context, task tracking, and dependency management. Triggers on "beads", "what should I work on", "session context", "track issue", "start", "start session".

Use when grep/ripgrep fails on Windows due to paths with backslash-space sequences creating malformed paths like 'nul' from \n+ul

Complete toolkit for web crawling and data extraction using Crawl4AI. 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.

This skill should be used when you need to manipulate the AoT DAG (Work Graph) structure. Use it when decomposing Atoms into sub-tasks, resolving Atoms, managing AND/OR dependencies, performing backtracking on OR branches, or shrinking the DAG by replacing resolved Atoms with summaries in Bindings.

Work with Internet Archive books and documents. Use when searching IA metadata, creating searchable indexes from OCR text, downloading pages, PDFs, or getting URLs, or querying local indexes. Triggers on "Internet Archive", "archive.org", "IA document", "OCR index", or historical book/document research.

Twenty CRM DevOps engineer specializing in Nx monorepo, Docker, CI/CD, and infrastructure. Use when configuring builds, optimizing pipelines, setting up Docker, managing deployments, or troubleshooting infrastructure.

Claude Codeのスキル（SKILL.md）を対話形式で作成するスキル。ユーザーが「スキルを作成」「新しいスキルを作りたい」「スキル作成を手伝って」と依頼した時に使用。個人用スキルまたはプロジェクトスキルを作成し、適切な構造とベストプラクティスに従ったSKILL.mdファイルを生成する

GitHub Issueの内容を分析して改修計画を自動作成するスキル。「issue XXを改修」「issueから改修して」「issue対応して」「issue番号XXを修正して」と依頼された時に使用。開発ワークフローに従ってブランチ作成から改修計画立案までを実行

ローカルブランチの状況を確認し、コミット済み・mainに反映済みの変更はブランチを自動削除してmainブランチを最新に更新するスキル。「ブランチを整理して」「ブランチクリーンアップ」「不要なブランチを削除」と依頼された時に使用。単一プロジェクトまたは全プロジェクト一括処理が可能

Supabase API endpoint creation. Use when creating API routes, database queries, or server actions.

Run verification commands and confirm output before claiming success

Systematically trace bugs backward through call stack to find original trigger

Run Playwright E2E tests for HEFAISTOS. Tests auth flows, navigation, page rendering, and responsive design.

Oracle Cloud Infrastructure MCP server providing comprehensive cloud management capabilities through the Model Context Protocol.

OCI Compute instance management capabilities

Helm chart authoring, values.yaml design, GitOps with Flux HelmRelease, External Secrets Operator integration. Use for creating and debugging Helm charts.

Use when creating technical presentations with Markdown, especially when embedding interactive JavaScript demos, iframes, or draw.io diagrams into reveal.js slides

Measure GEMV kernel relative error vs FP32 reference.

作業が完了/修正済み/通過済みと主張する直前、コミットや PR 作成前に使用する。成功の主張をする前に検証コマンドを実行し出力を確認することを要求する。常に主張より証拠

Post content to AI agent social network. Every 4 hours.

Scans web projects, analyzes page context with NLP, and inserts Unsplash hotlink URLs directly into HTML/JSX. NEVER downloads images locally - always uses Unsplash CDN hotlinks.

从预设知识库中检索答案，支持多轮追问和上下文理解。

Run Claude Code programmatically in headless mode or via bidirectional stream-json protocol. Use when building agents, automating Claude, running headless Claude, implementing multi-turn conversations, or integrating Claude CLI into applications.

Use when user explicitly delegates to Codex (@codex, ask codex, let codex, 问codex, 让codex, 请codex). Synchronous - waits for reply.

Scrape web pages and extract content via Firecrawl MCP

Search library documentation and code examples via Nia

Fast codebase search via WarpGrep (20x faster than grep)

Implement production-hardened JWT authentication for FastAPI + Better Auth applications. Use when implementing JWT middleware, fixing Windows encoding issues (Errno 22), resolving PostgreSQL schema mismatches, and enforcing user isolation in database queries. Handles stateless JWT auth flow with cross-platform stability and comprehensive error handling.

Browser automation with persistent page state. Use when users ask to navigate websites, fill forms, take screenshots, extract web data, test web apps, or automate browser workflows. Trigger phrases include "go to [url]", "click on", "fill out the form", "take a screenshot", "scrape", "automate", "test the website", "log into", or any browser interaction request.

Guide Test-Driven Development following Kent Beck's Red-Green-Refactor cycle and Tidy First principles. Use when developing features with tests, following Kotest BehaviorSpec with Korean Given-When-Then patterns. Covers Domain (pure logic), Application (UseCase with MockK), and Adapter (Controller/Repository) layer testing.

Create a new domain module following DDD + Hexagonal Architecture patterns for the blog-server project. Use when creating new modules in modules/ directory with domain, application, and adapter layers. Generates Aggregate Roots, Value Objects, UseCases, Services, Controllers, JPA entities, and Mappers following established patterns from the member module. All code includes Korean KDoc comments.

Better Auth TypeScript/JavaScript authentication library. Use when implementing auth in Next.js, React, Express, or any TypeScript project. Covers email/password, OAuth, JWT, sessions, 2FA, magic links, social login with Next.js 16 proxy.ts patterns.

Interact with Linear task management via `cg` alias — spawns a one-shot Claude Code session with Linear MCP connected.

Docker containerization patterns for production deployments. Covers multi-stage builds, image optimization, security best practices, and application-specific patterns for Next.js and Python/FastAPI.

This skill validates Dapr configurations for security, correctness, and best practices. It ensures all Dapr components and sidecar annotations follow standards before deployment.

Generate comprehensive Product Requirement Documents (PRDs) from rough ideas, feature requests, or informal specifications. Use when the user provides a high-level concept, informal requirements, or asks to "create a PRD", "write a spec", "document requirements", or needs to structure informal ideas into a formal PRD. Converts unstructured input into well-structured PRDs with clear goals, requirements, success metrics, and implementation phases.

Validates test suite quality through mutation testing. Generates intelligent code mutations, runs tests to verify they catch the changes, and identifies gaps in test coverage. Use when evaluating test effectiveness, validating newly written tests, or improving test quality for mission-critical code.

Guidance for nlsh remote mode (--remote flag). Use when commands execute on a remote server via SSH tunnel. Teaches: (1) Commands run on REMOTE not local, (2) Use upload_file for local-to-remote file transfers (not scp/rsync), (3) Use download_file for remote-to-local transfers, (4) Local env vars like NLSH_REMOTE_HOST are unavailable on remote.

Run backend integration tests for HTTP API endpoints and WebSocket events. Use when testing full request/response flows, API behavior, Socket.IO events, or end-to-end game scenarios.

Invoke this skill when: - Setting up authentication in FastAPI backend - User mentions "JWT", "token verification", or "auth middleware" - Phase II requires protected API endpoints

Invoke this skill when: - Setting up a new FastAPI backend with database - User mentions "Neon", "PostgreSQL", or "database connection" - Phase II implementation requires DB setup