AI Skills Hub - Browse 180,000+ Agent Skills

一个全面的智能体技能集合，涵盖上下文工程、多智能体架构和生产级智能体系统。在构建、优化或调试需要有效上下文管理的智能体系统时使用。

Long-term memory system for Claude Code using HelixDB graph-vector database. Store and retrieve facts, preferences, context, and relationships across sessions using semantic search, reasoning chains, and time-window filtering.

Image manipulation tool for resizing, rotation, flipping, cropping, padding, format conversion (JPEG/PNG/WebP/TIFF/HEIC), transparency operations (remove/replace/extract/blend), grayscale conversion, auto-cropping borders, and file size optimization. Use when users need to transform, convert, or optimize images.

Enforces project caching conventions when implementing cache layers using React cache(), Next.js unstable_cache, Upstash Redis, and Cloudinary. This skill ensures consistent patterns for cache keys, tags, TTL configuration, cache invalidation, and domain-specific CacheService helpers.

Enforces project Cloudinary media conventions when working with image uploads, URL generation, transformations, social sharing images, and photo management components. This skill covers the complete media workflow including CloudinaryService operations, URL utilities, CldImage/CldUploadWidget components, path organization, type definitions, and Sentry integration.

下载YouTube视频，支持自定义画质与格式选项。当用户请求下载、保存或获取YouTube视频时，可使用此功能。支持多种画质设置（最佳、1080p、720p、480p、360p）、多种格式（mp4、webm、mkv）以及仅音频下载为MP3。

本书提供专家级指导，帮助读者遵循现代最佳实践编写模块化、可投入生产的Bash脚本。当用户需要执行以下任务时，本书将派上用场：（1）编写新的Bash脚本；（2）改进或重构现有的shell脚本；（3）为脚本添加错误处理和适当的结构；（4）遵循可维护shell脚本的最佳实践；（5）调试或修复Bash脚本问题。

Generate speech from text (TTS), transcribe audio to text (STT), and run speech-to-speech pipelines using MLX on Apple Silicon. Use when working with audio generation, voice synthesis, speech transcription, voice cloning, or audio processing on Mac with MLX framework.

Feishu cloud storage file management. Activate when user mentions cloud space, folders, drive.

Local E2E debug and test framework for clawd-feishu plugin development. Use when debugging message flow, testing bot responses, verifying Feishu web UI interactions, or performing end-to-end validation of the OpenClaw-Feishu integration during development.

Set physical size of SVG for plotters/printers

Reduce number of colors in SVG by merging similar colors

Apply geometric transforms to existing SVG elements

Split SVG into separate files by color for multi-pen plotters

Convert all colors in SVG to a single color

Change the stacking order (z-index) of SVG elements

Duplicate existing SVG elements with optional offset and modifications

Remove SVG definitions (gradients, filters, patterns, masks, etc.)

Python package manager and project tooling using uv. Use when working with Python projects, managing dependencies, creating virtual environments, running Python scripts, configuring workspaces/monorepos, or troubleshooting uv issues. Triggers on pyproject.toml projects, uv commands, pip replacement workflows, Python version management, workspace configuration, or CI/CD Python setup.

Fetches web pages and performs web searches using local Chrome browser. Use when WebSearch or WebFetch tools fail with HTTP errors (403, 404, proxy blocks), when accessing sites behind corporate firewalls, or when the user needs to search using their authenticated browser sessions.

Constructs Google Ads Query Language (GAQL) queries with proper syntax, field selection, and filtering. Use when building custom Google Ads reports, exploring available fields, or troubleshooting GAQL syntax.

Interactive planning and execution for complex tasks. Use when user asks to use or invoke planner skill.

Analyzes deal state (stage + methodology) to recommend 3-5 specific, calendar-worthy next actions. Flags missing information and suggests which skills to run next. Use when the user needs tactical action recommendations to drive deal momentum forward.

Generate customized ROI models for sales opportunities by populating Excel templates with deal-specific cost savings, investment data, and financial projections. Use when the user requests an ROI analysis, business case, or financial justification for a deal.

Figma 디자인 파일에서 정보를 가져옵니다. 파일 구조, 노드 정보, 스타일 정보를 조회할 때 사용합니다.

Complete workflow generation - orchestrates internal, beta, and production deployment workflows

Create and run validation script to test Play Store API connection

PROACTIVELY identify and link relevant slipbox notes into narrative project structural documents. Use automatically when working on narrative projects to surface relevant research and arguments. Maintains project-specific tracking of which slipnotes are integrated where.

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.

PROACTIVELY use for Obsidian vault operations. Fast search using ripgrep, direct file operations (read/write/delete). Works with slipbox in _Slipbox directory. 4-7x faster than API. Use obsidian-api skill ONLY for Dataview queries, periodic notes, or patch operations.

Read and extract text from PDF files, Word documents, Excel spreadsheets, PowerPoint presentations, and images. Use when user asks to read, analyze, or extract content from .pdf, .docx, .xlsx, .pptx files or image files. Use markitdown to convert documents. Never use Read tool directly on binary files.

Debug AILANG code errors. Use when you encounter type errors, parse errors, or runtime failures in AILANG programs.

BigQuery MCP ツールのエキスパート。データ探索、スキーマ確認、SQL クエリ実行に使用。BigQuery のデータセット、テーブル、分析クエリの実行時にトリガー。

Debug failed or erroneous Dagster jobs through GraphQL API queries, event log analysis, and systematic troubleshooting. Use when investigating Dagster run failures, asset materialization errors, or pipeline execution issues.

Android security research and penetration testing. Use when analyzing APKs with apkleaks, testing API key permissions, bypassing security controls (root/Frida/SSL), hooking runtime methods, or capturing secrets. Follows OWASP MASTG methodology.

Guide for creating new skills in Kai's personal AI infrastructure. Use when user wants to create, update, or structure a new skill that extends capabilities with specialized knowledge, workflows, or tool integrations. Follows both Anthropic skill standards and PAI-specific patterns.

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).

Use when starting new Act project (CLAUDE.md doesn't exist), adding cast to existing Act (CLAUDE.md exists), or facing complex cast needing sub-cast extraction (>10 nodes) - guides through interactive questioning (one question at a time) from requirements to validated architecture with mermaid diagrams, emphasizing design before implementation, no code generation

Use when creating new cast package, installing/managing dependencies (monorepo or cast-level), resolving dependency conflicts or packages out of sync, or launching langgraph dev server - checks CLAUDE.md first for context, then handles all uv-based project setup and package management (dev/test/lint groups)

Maintain and manage AI-readable project metadata in a per-repo `.project/` folder (project.json + overview.md) with scripts for init/read/update/validate/format/render and incremental updates based on git commits since updated_at. Use when Codex needs to create/update `.project` info or answer questions about a repository’s overview/stack/architecture (including Chinese requests like “项目信息/项目概况/技术栈/项目架构/根据 git 提交更新”).

Comprehensive end-to-end testing automation for the Kosmos autonomous AI scientist project. Supports local models (Ollama), external APIs (Anthropic/OpenAI), and Docker sandbox for full workflow testing.

Build self-hosted AI chat assistants using CopilotKit + LangGraph. Use when implementing conversational AI interfaces with agentic backends, streaming responses, shared state between frontend and backend, or generative UI. This pattern uses NO hosted services - both CopilotKit runtime and LangGraph agent run on your own infrastructure. Triggers on requests to build chat interfaces, AI assistants, conversational agents, or integrate LangGraph with React frontends.

Build motions and appellate briefs from user-supplied DOCX templates using JSON or XML data. Preserves user formatting; requires template with {{placeholders}}.

Assemble FRAP 28-compliant Ninth Circuit opening briefs by copying user-provided sections into a fixed template/ordering. Never rewrite substantive text.

Legal Document Formatter for Pro Se Litigants. Formats ANY legal document using taxonomy + jurisdiction profiles. Read PimpJuice_instructions/MODEL_INSTRUCTIONS.md first.

Generates Ninth Circuit Court of Appeals cover pages using a perfect template. Use when Tyler needs to create a cover page for an appellate brief, motion, or other filing.

Interactive Ninth Circuit appellate brief formatter. Use when Tyler needs to format legal briefs for the Ninth Circuit Court of Appeals with proper styles, spacing, and optional content review with tracked change suggestions.

Use when the user wants to theme an application with Rose Pine colors - navigates Rose Pine documentation to find app-specific configuration and guides the user through installation

Use when configuring Niri window manager, writing niri config files, setting up keybindings, window rules, animations, or troubleshooting Niri issues - navigates comprehensive reference documentation

Guide for building AI agents in Go using adk-go framework. Use when creating agents with Gemini or OpenAI models, implementing tools, running agents with streaming, or managing conversation sessions. Triggers: Go agent development, adk-go, llmagent, google.golang.org/adk.

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

Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

Use when starting feature work that needs isolation from current workspace or before executing implementation plans - creates isolated git worktrees with smart directory selection and safety verification

Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies

Remote-control tmux sessions for interactive CLIs by sending keystrokes and scraping pane output.

Manage Apple Reminders via the `remindctl` CLI on macOS (list, add, edit, complete, delete). Supports lists, date filters, and JSON/plain output.

Manage Things 3 via the `things` CLI on macOS (add/update projects+todos via URL scheme; read/search/list from the local Things database). Use when a user asks AeonSage to add a task to Things, list inbox/today/upcoming, search tasks, or inspect projects/areas/tags.

Set up and use 1Password CLI (op). Use when installing the CLI, enabling desktop app integration, signing in (single or multi-account), or reading/injecting/running secrets via op.

# Canvas Skill

CLI to manage emails via IMAP/SMTP. Use `himalaya` to list, read, write, reply, forward, search, and organize emails from the terminal. Supports multiple accounts and message composition with MML (MIME Meta Language).

Kill dev server, regenerate Prisma client, push schema, and restart. Activated when the user says "restart dev", "kill and rerun", "refresh prisma", or after schema changes.

Write consistent, debuggable error messages for API routes and server actions. Activated when writing error handling code.

Write consistent toast notifications using Sonner. Activated when adding user-facing feedback messages.

ビジネスアイデア発想フレームワーク。Vehicle（手段・媒体）× Persona（人物）× Pain（課題）の3要素を組み合わせて革新的なアイデアを生成する。「ビジネスアイデアを考えたい」「新規事業のアイデア」「3-Pillarフレームワーク」「Vehicle × Persona × Pain」などと言及した時に使用。あらゆる業界・領域に適用可能。

デモのストーリーとシナリオを設計する。ハッカソンのデモ発表を準備するとき、製品の価値を体験として伝えたいとき、「何をどの順番で見せるか」を決めたいときに使用。

精准后期合成 Skill。支持长文本分句 TTS 合成、多行文本水平/垂直双中心对齐、多视频素材自动循环拼接。

This skill should be used when the user asks about "meeting notes", "what did we discuss", "find meeting", "show meeting", "recent meetings", "meeting transcript", "export meeting", "search meetings", "who was in the meeting", "yesterday's meeting", "last week's calls", or mentions Granola, meeting summaries, or wants to access meeting history. Provides access to Granola meeting notes, transcripts, and summaries.

Assembles coder and verifier prompts from templates. Use when user says "build a prompt", "assemble a prompt", "assemble prompts", or "generate a prompt" for dispatching work to coding and verification agents.

Use this skill when starting a new project or adding SynthesisFlow to an existing project. Scaffolds the directory structure (docs/specs, docs/changes) and configuration files needed for the spec-driven development workflow.

Use this skill when proposing new features or changes via the Spec PR process. Manages the creation, refinement, and approval of feature specifications before any code is written. Triggers include "create spec", "propose change", "start spec PR", or beginning feature definition.

Use this skill for early-stage project planning through Product Requirements Documents (PRDs). Guides users from initial project ideas through product briefs, market research, PRD creation, validation, and epic decomposition. Triggers include "create PRD", "product brief", "validate requirements", or beginning project inception activities.

Before doing anything else, run the installation checker:

Fetch unresolved PR review comments using GitHub GraphQL API, filtering out resolved and outdated feedback.

Check test status and debug failing tests using Wallaby.js real-time test results. Use after making code changes to verify tests pass, when checking if tests are failing, debugging test errors, analyzing assertions, inspecting runtime values, checking coverage, updating snapshots, or when user mentions Wallaby, tests, coverage, or test status.

Write and revise Markdown in a clean, consistent style. Draft and edit Markdown documents. Improve formatting for headings, lists, code blocks, tables, or links.

Four-phase debugging methodology emphasizing root cause analysis before fixes. Use for bug investigation, preventing random fixes, and systematic problem-solving. Based on obra/superpowers.

Multi-repository git synchronization patterns for batch operations

Generate comprehensive, maintainable test suites through code analysis. Use when the user asks to write tests, create unit tests, add test cases, test a function, or improve test coverage.

macOS launcher automation with Raycast extensions (TypeScript/React) and Alfred workflows (AppleScript/Python) for keyboard-driven productivity

pdf-large-reader Skill

Deep learning for single-cell analysis using scvi-tools: data integration, batch correction, multi-modal analysis, reference mapping, and more.

Capture, organize, and surface institutional knowledge across sessions and repositories

Synchronize skills from anthropics/knowledge-work-plugins into workspace-hub skill system

Write, update, and troubleshoot devctl plugins that speak the NDJSON stdio protocol v2 (handshake + request/response/event frames). Use when creating a new devctl plugin, converting repo scripts into devctl pipeline ops, adding dynamic commands, wiring .devctl.yaml, or debugging protocol contamination/timeouts and other plugin failures.

Schema-first workflow for sharing protobuf-defined payloads between Go and TypeScript with JSON transport, Buf codegen, protojson, and @bufbuild/protobuf. Use when creating proto schemas, generating Go/TS code, emitting JSON payloads, decoding with fromJson, handling int64/Struct, or validating end-to-end payloads.

Write and maintain an implementation diary: step-by-step narrative of what changed, why, what worked, what failed (with exact errors/commands), what was tricky, and how to review/validate. Use when a user asks to create/update a “Diary” doc, record work for a ticket, document debugging outcomes, or produce code review instructions and future follow-ups.

Documentation management with the `docmgr` CLI: create and work in ticket workspaces (`ttmp/...`), add documents, relate code/files to docs, manage tasks/changelogs/metadata/vocabulary, and search/validate docs. Use when a user mentions `docmgr`, ticket docs, `docmgr doc relate`, `docmgr doc search`, YAML frontmatter validation, or asks to turn ad-hoc markdown into a structured, searchable knowledge base.

Generate and update translation files (.pot) for WordPress i18n. Use before creating PRs to ensure all translatable strings are captured. Required workflow before any PR submission.

GitHub CLI commands for workflows, PRs, and issues. Use when checking CI status, monitoring workflows, or interacting with GitHub. CRITICAL - always append '| cat' to prevent pager hang.

Cut a release (bump version, commit, tag, push)

Proxmox VE sysadmin for home lab infrastructure. Use when triaging services, checking container/VM status, viewing logs, managing resources, or debugging self-hosted apps. Can discover infrastructure dynamically via SSH.

Convert mermaid diagrams in markdown files to PNG images. Use when the user wants to export markdown with mermaid to formats that don't support mermaid (Google Docs, PDF, etc).

Project architecture, coding conventions, and development standards. Use when creating files, modifying package.json/project.json, writing tests, making architectural decisions, or when a broad overview of the monorepo would be useful.

Development skill for rohanverma.net Zola static site. Use when: editing styles, fixing responsive issues, adding pages, modifying templates. Keywords: zola, scss, css, mobile, responsive, template, design, theme, blog, project.

Forensic audit of client social media history with engagement metrics, trend analysis, and strategic recommendations. This skill should be used when (1) onboarding a new client to assess current social performance, (2) quarterly or annual performance reviews, (3) before creating a new social strategy, (4) client asks about what's working/not working on social. Requires CSV/Excel exports from Instagram, Facebook, GBP in per-channel folders.

Guide for creating effective Claude Skills. Use when (1) creating a new skill from scratch, (2) updating an existing skill's structure or content, (3) packaging a skill for distribution, (4) learning skill design patterns. Requires Python 3.x for automation scripts.

Run .NET architecture linter to check for MSBuild/CPM violations

Prevents shortcuts and cheating when completing tasks. Blocks commits with warning suppressions, commented tests, or deleted assertions.

Package entire code repositories into single AI-friendly files using Repomix. Capabilities include pack codebases with customizable include/exclude patterns, generate multiple output formats (XML, Markdown, plain text), preserve file structure and context, optimize for AI consumption with token counting, filter by file types and directories, add custom headers and summaries. Use when packaging codebases for AI analysis, creating repository snapshots for LLM context, analyzing third-party libraries, preparing for security audits, generating documentation context, or evaluating unfamiliar codebases.